Project and Program Risk Management by slbsema

VIEWS: 959 PAGES: 57

									Project and Program
UNIVEE31DAD DE BELGhtAF4G Geiito d e Esrlidios de Poçyrado y Edt~cacióri Zontrnua

RISK MANAGEMENT
A Guide to Managing Project Risks and Opportunities

Edited By: R. Max Wideman Fellow, PMI

A Publication of the Project Management Institute Four Campus Boulevard Newtown Square, Pennsylvania 19073USA 610/356-4600 Fax: 610/356-4647 E-mail: pmihq@pmi.org Web: http:/,www.pmi.org

Library of Congress Cataloging-in-PublicationData Edited by: Wideman, R. Max. Project and program risk management: a guide to managing project risks and opportunities/R. Max Wideman p. cm .-(The PMBOK handbook series: v.no.6) Includes bibliographical references. ISBN 1-880410-00-1 (10 vol. set).-ISBN 1-880410-06-0 (vo1.6) 1. Industrial project management. 2. Risk management. I. Title. 1 . Series. 1 HD69.P76W54 1992 658.4'04-dc20

92-3336 CIP

ISBN 1-880410-06-0 (Volume 6) 1-880410-00-1 (10 Volume Set)

1 Copyright 0 1 9 9 2 by the Project Management Institute. A 1 rights reserved. Printed in the United States of America. No part of this work may be reproduced or transmitted in any form or by any means, electronic, manual, photocopying, recording, or by any information storage and retrieval system, without prior written permission of the publisher. "PMI" is a federally registered trade and service mark; "PMP" and the PMP logo are federally registered certification marks; and the PMI logo, "PMBOK" and "Building professionalism in project management." are trademarks of Project Management Institute. PMI books are available at quantity discounts. For more information, please write to the Business Manager, PMI Publishing Division, Forty Colonial Square, Sylva, North Carolina, 28779 USA. Or contact your local bookstore. The paper used in this book complies with the Permanent Paper Standard issued by the National Information Standards Organization (239.48-1984).

This publication is the second in a series of nine handbooks developed by the koject Management Institute (PMI). The senes is designed to complement the Project Management Body of Knowledge (PMBOK) as codified by PMI. This handbook deals with the function of Risk Management in the context of project or program management. It establishes the need for exarnining risks and opportunities in project work and describes approaches which can lead to significant reduction of the risks as well as to better project performance. This is a sometimes overlooked aspect of project management which can often result in significant improvements in the ultimate success of the project. This handbook should appeal equaiiy to those who are looking for a sumrnary of the basics of project and program risk management, as well as to those who are studying for PMI's certification program. In addition, the material in the handbook will benefit course developers, trainers and academics who are seeking guidance as to appropriate project and program risk management educational content. In keeping with previous practice, this first printing of the handbook is being made available as a "Preliminary Issue for Trial Use and Comment." Since the content is tied specifically to the PMBOK, it is important to ensure that PMI members support the material and its manner of presentation. I hope, therefore, that readers will take a moment to contribute constructive suggestionsfor improving and upgrading the next issue by completing and retuming the Comments page at the end of this book. The material in this handbook has been assembled entirely by voluntary PMI member effort and,as anyone knows who has been involved in similar undertakings, requires many hours of dedicated effort. In particular I would like to acknowledge the efforts of Max Wideman and the many people who provided comments and suggestions on the vanous drafts. To all those who contributed their thoughts and writings to this first issue, I express our appreciation for a job well done! Rodney J. Dawson President, Project Management Institute

Preface
Welcome to project and program risk management! 1n a very real sense, progress is made by spotting opportunities and taking advantageof them. Once identified, the most aggressiveway of taking advantage of such opportunities is by establishinga projed. Thus, the whole point of a project is to introduce some change which wiU be beneficial in finanaal or other ways to the sponsors and users of the project. However, introducing change m a introducing something new, to venture forth, to take some risks. So risks are the corollary of opportunity. They represent exposure to mischance, hazards, and the possibility of adverse consequences.They are the down-side of a proyxt undertaking. While pro* managernentis the best way of m a n a p g opportunity, taking risks has always been a fundamental part of the process, and this needs to be pro-actively managed through the function of Pro@ and Program Risk Management. Generally, when we speak of taking a risk we tend to think of those things which are at long odds, are highly chancy and possibly hazardous. Yet many risks we take in everyday life are so comrnonplace that we scarcely give them a thought. Instead, we treat them as mere uncertainties and react to them subconsaously, taking precautions that experiencehas taught us are only prudent. However, in today's markets with more difficult economic conditions, tougher competition, and ever advancing technology, project uncertainty and risk have assumed significantly greater proportions. Indeed, in most projects, not oniy are the uncertainties and risks numerous, but they are &o interrelated. This affects project resuits in complex ways, making it difficuit for management to be confident in forecasting the final results. Therefore, Project and Program Risk Management is seen as the formal process whereby the risks and opportunities are systematically identified, assessed and appropriately provided for in the course of project planning and implementation. It means taking a pro-active stance to cultivate an environment in which project and program risks are significantlyreduced, if not eliminated entirely, and opportunities are cultivated. Project and Program Risk Management should encompassthe fui1spectrum of activities associated with the handling of project uncertainties. As one of project management's integrative functions (see PMBOK Handbook Volume 1, A Framework for Projecf and Program Management Integratwn) Project and Program Risk Management is inextricably tied into each of the other project management functions, especially the four basic Time project constraints of Scope, Q ~ a i i ~ ; and Cost. It is, therefore, a key function of the project management process. However, management's attitude towards risk, in many cases, is governed by the extent of their understanding of the risk management process, their confidente in the associated techniquesand in the analytical results obtained. Others consider the subject too mathematical, yet many risks relate to people and their attitudes. Hence the need for ti handbook hs

Acknowledgments
Understandably, the risks associated with large and very large projects have received noteworthy attention in the literature, because of the substantial sums of money involved on the one hand, and the political and professional reputations of the sponsors on the other. On lesser projects, however, aspects of the project which may be at risk are frequently submerged in an ali-encompassing "contingency ailowance" determined as a result of some previous experience. For many, this approach may be quite satisfactory, yet closer examination of the subject indicates that including risk as a pro-active management function provides project management with an added management opportunity. Project and Program Risk Management furnishes the chance to better understand the nature of the project at hand, to involve team members in its strengths and weaknesses, and generaliy to integrate thecore functions of Scope, Quality, Time and Cost with the interactive functions of Hurnan Resources, Contract/Procurement, and lnformation/Communications management. For these reasons, PMI early recognized the importante of this subject to the overali success of a project, no matter what its size or area of application, and so identified Project and Program Risk Management as a separate PMBOK topic. Once again many PMI mernbers, too numerous to cite ali by name, have enthusiasticaliy volunteered material, suggestions and encouragement for the development of this handbook. I would, however, like to mention a few who have been particular contributors or my mentors in this effort: Mike Curran, David Hamburger, and Bili Hosley for most helpful explanations, conversations and text; David Hulett for a copy of an excellent Orange County Chapter workshop handout on Risk Management; Pat Buckley and Biil Duncan for a change in emphasis; Chris Quaife for both written contributionsand thorough and perceptive comments on my original texts; and Davidson Frame for encouragement and testing the handbook in one of his classes. I am sure that readers will benefit greatly from their work. Any errors of interpretation or omissions are, of course, my own responsibility. As always, my wife Audrey has been most supportive of this work and patiently endured through many hours of moans and groans arising from the effort of putting pen to paper (or more conternporarily, fingers to keyboard). R. Max Wideman, Fellow, PMI Editor

iii

Contents
Foreword . . Preface . . Acknowledgments
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Chapter VI
. . . . . . Response Options . . . . . . System Standards Insurance . . . . . . . . Response Planning . . . . . . Data Collection, Application and Documentation

Risk Response and Documentation
VI .l VI .l VI . 2 VI-3 v1-3

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

...
111

I II

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . .

Chapter I Introduction
A.

i. 3
C. D E. F.

Handbook Purpose and Content . Why Project Risk Management? . Uncertainty, Opportunity and Risk The Nature of Risk Management . Project Risk Management is Pro-active Risk and Decision Makers . .

. .
.

. .
.

. .
.

. .
.

. .
.

. .
.

. .
.

. .
.

. .
.

. .
.

. .
.

.
.

. . .

. . .

. . .

. . .

. . .

. . .

. . .

. . .

. . .

. . .

1-1 1-1 1-3 1-4 1-4 1-5

Chapter VI1 Management of Contingency Allowances
After the Known Risks. What Then? . . . Contingency Reserves? . . . . . . Application of Project Contingency Allowances . Contingency Allowances for Project Implementation Implementation Contingency Strategie . . A Simple Tabular Calculatio . . . . .

. . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. .

. .

. .

. .

. .

. .

. .

. .

VI1 .1 VI1 .2 VI1 .3 VI1 .3 v11 .4 v11 .6

Chapter II Integration. General Approach and Definitions
Integrating Risk into Project Management . . . The Natural Risk Management Sequence . . . Risk Management .An Integrative Function . . . . . . Project Risk Management Definition Variation of Risk Factors Through the Project Life Cycle . . . . . . . Four-Phase Approach . . . . . . . . . . . . . . . . . . . . .
II .1 II .1 11-3 11 .3 11 .3 11-6

Chapter Vlll

Managing the Risks of the Project's Environment
. . . . . . . . . . . . . . . . . . . . . . .
VIII .1 VIII .2 VIII-3 VIII .5 VIII .5 V111- 6

.
.

.

.

.

.

.

.

.
.

.
.

.
.

.
.

.
.

.
.

What is the Project Environment? . . . . . . . . . . . Problems are Caused by People Risk and Safety . . . . . . . . Principal Deteminants . . . . . . . Managing by Stakeholder Groupings and Categones . . The Means and Value of Exercising Positive Influence Contract Strategy Considerations Suggested Risk Sharing Principles Types of Contracl . . . Different Contract Risk Implications More than One Contract . . A Question of Attitudes . .

. . . . . .

Chapter III
Risk in Corporate Business Management . . Risk in Project Management . . . . . Types of Risk . . . . . . . . . Project Risk Identification . . . . . Project Risk Configuraiion . . . . . Risk Factors . . . . . . . .

Risk Identification
.
. . .
111 .1 111 .1 111 .2 111 .3 111 .4 111 .6

. . . . . . . . . . . . . . . .
.
. . . . . .

Chapter IX
.

Dealing with Risks in Contracts
IX .l IX . 2 IX . 3 IX .4 IX .4 IX-6

.
. . .

.

.
. . .

.
. . .

.
. . .

.
. . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Chapter IV Risk Assessment Goals and Methodology
Risk Assessment Goals . . . . Assessment Methodology . . . Advantages of Assessment Methodology The Basics of Probability . . . The Quality Risk . . . . . . . . . . The Schedule Risk

. . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . _ . .

. . . . . . . . . . . .

. . . .
.

.

IV .1 IV .2 IV .6 IV .6 IV-9 IV-9

Risk Management .The Present It's the Attitude That Counts . Risk Management .The Future

. . . . . .

. . .

Chapter X SummarylConclusions . . . . . . . . . X- 1 . . . . . . . . . X-2 . . . . . . . . . X.2

Chapter V
Data Storage. Retrieval and Computation . . PERT and the Probabilistic Model . . . Range Estimating . . . . . . . Risk Analysis. . . . . . . . . Knowledge-Based Risk Management . . .

Computer Applications
V-1 V- 1 V-2 V-4 V-5

. . . . . . . . . .

. . . . .

. . . . .

. . . . .

. . . . . . . . . . . . . . .

Appendices
FIGURE

Illustrations
Page
1-2 The Uncertainty Spectrum . . . . . . . . . . . . . 1-3 TheUncertaintylOpporhinitylRiskRelationship . . . . . . . . . 11-2 Integrating Risk With Other Project Management Functions . . . . . . 11-5 Typical Life Cycle Profiles - Risk vs. Amount at Stake . . . . . . . 11-6 Risk Function Breakdown: Four Processes . . . . . . . . . 111 - 4 Risk Management: Identification . . . . . . . . . . . . 111 - 5 Risk Management: Mitigation . . . . . . . . . . . . iv-3 Companng Severity of Project Risks . . . . . . . . . - . IV - 4 Impact Analysis Matrix Sequence . _ . . _ . . . . . . IV - 10 The Quality Risk: Difference Between Success and Failure . . . . . . IV-11 Simple Example of Network Completion Risk . . . . . . . . . v-3 Probability Distributions: Two Projects With the Same Expected Values of Total Costs v-4 Graphical Presentation of Analysis Results . . . . . . . . . v-5 Presenting Results of Anlaysis: Cumulative Risks vs. Cnterion Value - ConstnictionProject V-6 . . . . . . . . . . Breakdown Structure of Identified Risks Project Environment Process . . . _ . _ . _ . . . . VIII - 1 VIII - 2 Project Environment: Externa1 Influentes and Interfaces . . . . . . VI11 - 7 Managingthe Environment: Public Relations Concept . . . . . . . IX - 2 Contract Type vs. Risk Allocation . . . . . . . . . . . . IX - 4 Scope Definition - Risk - Contract Selection . . . . . . . . . B-5 Conceptual Relationship Between Risks, Damage Scenanos and Consequences . 8-7 Simplified Risk Analysis Program Structure . . . . . . . . . C-4 Decision Tree for Two Projects Showing Probabilities Assigned . . . . . D-4 Comparing Severety of Project Risks . . . . . . . . . . .

A. Typical Project Risks
How Does the Project Manager Know When There is a Project Risk? Specific Project Risks . . . . . . . . . .

.
. .

.
.

. A-1 . A- 1

B. Impact Analysis Methodology
Problem Structuring . . . Risk Identification and Screening . . . Risk Quantification Risk Combination and Modeling . . . Overall Evaluation Project Risk Report . . .

. .
.

.

. .
. . .

. .
. .

. . .

. .

. . . . .
.

.

. . . . . . . . .

. .

.
. .

.
. .

. .

. .
.

.
. . .

. . . . . .

.
.

. . . . .
.

. B-2

.

. B-4 . B-6 . B-6

. B-4

. B-3

C. Other Risk Analysis Techniques
Brainstorming . . Sensitivity Analysis . . Probability Analysis . . Delphi Method . . Monte Carlo Decision Tree Analysis Utility Theory . . . Decision Theory
.

.
. . . . . .

. .

.
.

. . . .

.

.
. .

. .
. . . . .

.

. . . . . .
. .

. . . . .

. . . .
.

. . . . . .

. .

. .

. . . . . . . . . . . . . . . . . . . . . . .

. . .

. . . .

.

.

. . . .
.

.

. C- I

. .
.

. . . . . . . . . . . . .

.

C- 1 C-2 C-2 C-3 C-3 C-4 C-4

D. Risk Applied to Schedule and Cost E. A Glossary of Project and Program Risk Management Terminology

TABLE
11.1 IV.1. IV.2. VII.l V11.2 IX.l Typical Functional Distribution of Controllable Risk Items . . . Network Activity Summary . . . . . . . . . . . . . . . . . . . . Network Path Evaluation Contingency Allocation: Consbuction Project . . . . . . Simple Tabular Calculation of EstimatingContingency . . . . Risk Implicationsof Different Types of Contract (from Client's perspective)

Page

. . . .
.

. . . . .

. . .
.

.

. .

11 - 4 IV - 11 IV-11 VI1 - 5 VI1 - 6 IX - 5

vii

Risk Management

Chapter
A. Handbook Purpose and Content

Introduction

The purpose of this handbook is to provide a sirnplified understanding of the nature of project nsk and opportunity, and a systematic approach to risk reduction. For brevity and ease of reference, it is deliberatelystructured in simpleand concise t e m , often in bullet format, or by graphic illustrations. This handbook is one of a family of nine, designed to provide a summary leve1review of each of the major functional areas of project management. These functional areas are those codified in the Project Management Institute's Project Management Body of Knowledge (PMBOK).This handbook provides a useful insight into the process of project risk management for anyone contemplating a project, whether it is a small administrative project, a large capital works project, or anything in between. It alço provides an introduction for those intending to study for PMI's Project Management Professional (PMP) certification exam.

B. Why Project Risk Management?
Corporate management has the responsibility to make formal judgments and appropriate decisions that will lead the organization to a successful destiny. Ideally, such decisions should be taken in an environment of total certainty, wherein a11 the necessary information is available for making the right decision, and the outcome can be predicted with a high degree of confidence. In reality, most decisions are taken without complete information, and therefore give rise to some degree of uncertainty in the outcome. In the extreme case of complete absence of information, nothing is known about the outcome and total uncertainty prevails. Organizational survival in today's world is achieved by pursuing opportunity within this spectrum of uncertainty and projects are typically launched to take advantage of these opportunities. Thus, the whole point of undertaking a project is to achieve or estabiish something new, to venture, to take chances, so risk has always been an intrinsic part of project work However, in today's markets, with heavy competition, advanced technology and tough econornic conditions, risk taking has assumed significantly greater proportions.

Chapter I

Introduction
The goals of risk management, therefore, are to identify project risks and develop strategies which either significantly reduce them or take steps to avoid them altogether. At the same time, steps should be taken to maximize associated opportunities. In essence, it involves planning which rninirnizes the probabiiity and net effects of things going wrong, and carefully matches responsibiiity to residual risks which are unavoidably retained. It is a very constructive and creative process. As a simple example, one way of avoiding a possible traffic jam while driving on the highway to a particular destination is to consider alternative forms of transportation. Granted that each may have its own particular set of risks, but careful comparison should identify the best set of alternatives with the lowest overail degree of uncertainty or risk of amving late. However, the impact of each on the time and cost of the journey must also be taken into account if the best overali arrangement for a successful amval is to be achieved. The selection may weli depend on the relative priorities given to the cost, schedule, and quality of the journey! If the real objective of the exercise is to hold a meeting, then perhaps the opportunity could be taken to hold the meeting at a more favorable intermediate location? In short, the purpose of project risk management is to: Specificaily ident* factors that are likely to impact the project objectives of Scope, Quality, Time and Cost Quantify the likely impact of each factor Give a baseline for Project Noncontrollables Mitigate impacts by exercising influente over Project Controllables The scope for project risk management lies somewhere between the two extremes of total certainty and total uncertainty, as shown in Figure 1.1.

Risk Management

C. Uncertainty, Opportunity and Risk From the foregoing it will be seen that uncertainty, opportunity and risk are closely alIied. It can be visualized that unknowns about the future may tum out to be either favorable or unfavorable, but lack of knowledge of future events constitutes uncertainty so that uncertainty is simply the set of all possible outcomes, both favorable and unfavorable. In this relationship, the probability of those outcomes which are favorable may be viewed as opportunity, while the probabiiity of those outcomes which are unfavorable 1 represent risk. Similarly, most opportunities when pursued cany with them associated risks and, generally speaking, the greater the opportunity, the greater is the degree of uncertainty and the consequent associated risk. Thus, opportunity and risk are also tied together and, indeed, one may be seen as the corollary of the other. This relationship is shown diagrammatically in Figure 1.2.

VENTURE (Project)

OUTCOME (Products)

FAVORABLE (Opportunity) UNKNOWNS (Uncertainty)

u

i

UNFAVORABLE (Risks)

PARTIAL INFORMATION

COMPLETE INFORMATION

I I

(UNKNOWN UNKNOWNS)

(KNOWN UNKNOWNS)

(KNOWNS)

II
I

THE GOAL: Project Risk Management's function is to move uncertainty away from risk and towards opportunity
Figure 1.2. The UncertauiSl/Opportunity/Risk Relationship

TOTAL UNCERTAINTY

GENERAL UNCERTAINTY

SPECIFIC UNCERTAINTY

TOTAL CERTAINTY

In the context of project management Project Risk is defined as foilows:

SCOPE OF PROJECT RISK MANAGEMENT'

Project Risk is the cumuiafive @ed of the chances of uncerfain occurrences adversely affecting projecf objecfives.
In other words, it is the degree of exposure to negative events, and their probable consequences impacting on project objectives, as expressed in t e m of scope, quality, time and cosi.' However, justas risks are associated with pursuing opportunities, so opportunities also flow from encountering risks.Indeed, many a risk which occurs in the event can, with a little foresight

'Note: In this range the informationto be sought is known

Figure 1.1 The Uncertainty Spectnun

Chapter I

Introduction
and ingenuity, be turned into an opportunity! Unfortunately, aii too often risks are either ignored or dealt with in a very arbitrary way so that such opportunities are overlooked. The constant goal of project risk management should be to move uncertainty away from risk and towards opportunity. Consequently, when assessing overall impacts of uncertainty on a project, it is the net project risk which should be determined, i.e., the cumulative net effect of the chances of both adverse and favorable consequences affecting project objectives.

Risk Management
Indeed, the t e m Project Risk Managemenf itself tends to be misleading because management implies complete control of events. On the contrary, project risk management should be seen as advanced preparation for possible adverse future events, rather than responding as they happen. With such advanced planning it should be possible to select an alternative action plan which wiii still enable project objectives to be achieved successfdly. Consider this improbable, but quite possible, situation. You are at risk of being shot at. You have four options.
Reactive:
1. You can move to avoid the builet; 2. You can deflect the bullet; or 3. You can repair the damage done by the bullet.

D. The Nature of Risk Management Generally, when we speak of taking a risk we tend to think only of those things which are highly chancy or hazardous. Yet rnany risks are so commonplace in everyday life that we scarcely give them a thought. Instead, we react to them subconsaously, and take precautions that experiente has taught us are only prudent. In crossing the road, we take the precaution of looking both ways and only cross when the road is clear. If we are in a huny, we might "take a chance" (increase the risk) by crossing when we see a sufficient gap in the traffic. If the traffic is heavy, and the risk appears to be extreme, then we might walk further to a designated crossingarea at an intersection with traffic lights, or even to a safer overpass, if there is one (risk reduced by trading off time and energy). Rarely do we systematically ident* ali the risks involved in reaching our destination. Even less do we consider the consequences should our chances fail to come off except, perhaps, once a serious accident has actually occurred. Otherwise, we might decide never to go anywhere at all! When it comes to our own family, however, we are inclined to take an entirely different approach. After all, we are now d e a h g with something very precious and with a lot of potential. When our chiidren are small, we admonish them not to go near the road (risk identification and avoidance).When they must cross the road, say, to get to school, we examine the dangers and either teach them how to cross safely or direct them to the school crossing guard (risk assessment and planning or shift of responsibility). When they get home at the end of the day, we ask them "how they got on"?-perhaps we can do something more to help them for tomorrow (informationfeedback and corrective action). We also make a mental note for when our youngest reaches the same age (building the data base). Thus, we have established the basic elements of managing project risk. E. Project Risk Management is Pro-active Project rnanagers wili recognize the classic systems methodology outlined in the previous section. This consists of input, process, output and feedback loop, a basic model which is so vital to the effective control of any project. Yet risk is somehow different. lt has to do with uncertainty, probability or unpredictability, and contingent planning.

Pro-active:
4.

You can take steps to avoid being confronted by the person with the gun.

Crisis management (reactive mode) consists of selecting the appropriate response. However, if anticipation and planning make it possible to avoid the situation in the first place (pro-active mode) then this approach would obviously be better. Unless, of course, there are compelling reasons to the contrary such as conducting a controversial political campaign! The point is: at no time f are you in mntrol o the bullet (the risk e ~ e n t ) ~

F. Risk and Decision Makers A risk should only be taken when the potential benefit and chances of winning exceed the remedial cost of an unsuccessful decision and chances of losing by a satisfactory m a r p . Therefore 4 the risk taker should obtain realistic answers to questions such as:
Why should the risk be taken? What will be gained? What could be lost? What are the chances of success (and failure)? What can be done if the desired result is not achieved? 1s the potential reward worth the risk? Risk elements that tend to attract and/or determine the response attitudes of decision makers include:' Potential frequency of loss Arnount and reliability of information available Potential seventy of loss Manageability of the risk Vividness of the consequences Potentiai for (adverse) publicity Ability to measure the consequences Whose money is it

Chapter I

Introduction
Severity of the potential loss appears to attract the most attention because individuals appear to be willing to accept small (even frequent) losses, but are averse to a nsk which has high stakes. Even sol a major thrust must be to minimize unwarranted optimism, prejudice, ignorante or self-interest. Good ways to do this include responding to the following i ~ s u e s : ~ Has it been done before? How close is the analogy? and Seek out corroborative evidence. Get personal interviews with those with the expenence. Obtain alternative opinions. Insist on written assessments, quantitative ifpossible. Avoid creating additional risk by mshing, understaffing, underfunding, or ignoring the obvious. Simply try to adopt well-established good project management practices, and so escape the well-known adage: "There's never enough time to do it right the first time, but time enough to do it over if its wrong!" But sooner or later a decision has to be made. Never lose sight of the basic reason for taking a nsk-to gain a specific reward. Some typical project rewards (or benefits) include: Achieving a desired result using limited resources, Advancing the state-of-the-art, Meeting a required end date or improving a schedule, Enhancing profitability, lncreasing a budget or schedule contingency, Saving money or offsetting a fiscal vanance, Improving the fim's market position, or Ensuring customer satisfaction. In each case, the potenüai reward must be accurately defined and, if at aii possible, measured i .common tenns such as cost. If a finite dollar value cannot be assigned to a particular outcome (e.g., customer satisfaction), it may be necessary to assign an arbitrary value to the intangible benefit for purposes of risk/reward evaluation. This may be far more difficuit than it seems as rnarketing's valuation of customer satisfaction, for exarnple, will generally exceed a judgment made by rnanufacturing or engineering. Decision makers who use information from various sources in evaluating the risk must temper their b d p d g m e n t s to ensure objectivity in the Ço: risk assessment pro~ess.7 Make the best decision given the state of knowledge-even sol it may not work out. Distinguish between a good decision and a good outcome. Since it is not possible to be certain of a good outcome, increase the probability of good outcomes by making good decis'ions!

Risk Management

Here are some helpful rules-of-thumb for the project managec9 Donlt take the risk if: The organization cannot afford to lose; The exposure to the outcome is too great; The situation (or the project) is just not worth it; The odds are not in the project's favor; It is no more than a "fair bet"; The benefits are not identified; There appear to be a large number of acceptable altematives (The greater the number, the more the uncertainty.); The nsk does not achieve a project objective; The expected value from the baseline assumptions is negative or is negative with small changes in assumptions; The data is unorganized, without structure or pattem; There is not enough data to compute the results (Get more data or do research.); A contingency plan for recovery is not in place should the results prove to be less than satisfactory.

1. Construction Industry Institute, Management of Project Risks and Uncertainties, The University of Texas at Austin, October 1989. 2. PMBOK, 3/28/87, p E-2. n 3. D.C. Fraser, Risk Minimisation i Giant Projects, International Conferente on the Successful Accomplishment of Giant Projects, London, England, May 1978. 4. D. Hamburger, The Project Manager: Risk Taker and Contingency Planner, PMJ, June 1990, p44. 5. Construction Industry Institute, Management of Project Risks and Uncertainties, The University of Texas at Austin, October 1989, p3-4. 6 . J.N. Brooke, Leveraged Risk Reduction, Proceedings PMI Seminar/Symposiurn, Atlanta, Georgia 1989, p302. 7. D. Hamburger, The Project Manager: Risk Taker and Contingency Planner, PMJ, June 1990, p45. 8. D.T. Hulett, PMP Certification Workshop - Risk Management, PMI Orange County Chapter, 1991, p12. 9. Ibid., pl and p58.

Risk Management

Chapter II

Integration, General Approach and Definition

A. Integrating Risk into Project Management
Experience on many projects reveals poor performance in terms of reaching scope, quality, time and cost objectives. Many of these shortcomings are attributed either to unforeseen events, which might or might not have been anticipated by more experienced project management, or to foreseen events for which the risks were not fully accornmodated. Perhaps one of the biggest hurdles is that of managemeni's attitude to risk itself. Some have little understanding of the concepts, while others lack confidencein the mathematical techniques and results obtained, prefemng to rely alternatively on aggressive risk taking or undue caution. Or again, inherent risks may simply be optimistically ignored. In reaiity, far better decisions with higher chances of project success are reached by facing these issues. For some project managers this may represent a new working environment. Figure 11.1 shows schematicaliy how the function of project risk management is inexticably tied into the remaining seven PMBOK management functions. Note how specific areas of risk are associated with each of the individual functions as shswn on the diagram. Each should be carefuliy evaluated as part of the risk management responsibility. A long list of potential project risks is provided in Appendix A. Not a11 of these risks apply to all projects, of course, but many do, and failure to manage risk in the manner described in the following chapters can lead to sigmficant and unnecessary losses. Poor past track records should be significantly improved by a generaliy better understanding and appiication of the project risk management function.

B. The Natural Project Risk Management Sequence
The greatest degree of uncertainty about the future is encountered in the concept phase of a project. Directions taken by the project sponsors in this phase have the greatest influence on the ultimate scope, quaiity, time and cost of the project. Also, change is an inevitable part of the iterative nature of managing projects, yet its extent and effects are often under-estimated at this time. Therefore, the need for a process for the realistic appraisal of factors affecting the accomplishment phases of the project is essential.

Chapter II

Integration, General Approach and Definition

Risk Management
and prioritizing the areas on which risk management should be focused. A systematic approach is required to sort through the mynad of uncertainties, to pinpoint the truly critical ones, and to identify effective ways of reducing those uncertainties, consistent with overall project objectives. In practice, depending on the size and nature of the project, effective risk management may require some quite detailed quantitative assessment of the impacts of the various uncertainties. This data provides a basis for judging the reliability of the original estimates, the effectiveness of possible alternative strategies, and for planning the best overall responses.

I
SCOPE

(

PROJECT MANAGEMENT INTEGRATION

I
INFORMATIONI COMMUNICATIONS

Life Cycle and Environment Variables ~x~ectations

I

I
I

I

ounLITY

p q

J-

Ideas, ~irectives, Data Exchange Accuracy

e uirements~ P R ~ ~ ~ T ~ ~ v a i I a b i l ~ v + I Standards Productivity RESOURCE

I

Time Obiectives.

a
I

, ,e s

~~ant, ~ateria~s:

Cost Objectives, Restraints

I

CONTRACTI PROCUREMENT

I

COST

I
After C . Quaife, 1/11/90

C. Risk Management An Integrative Function As noted in Section A, failure to give proper recognition to risk management on a project can lead to unnecessary and often substantial losses, or even complete project failure. The status of risk on a project varies significantly during the course of its life cycle, and, as with most of the other project functions, the most effective time for achieving the greatest irnpact on project results is early on in the project development phase. Consequently, risk management should be established as a continuing integrative function throughout the project's life cycle. Figure 11.1 iilustrated schematically how nsk management integrates with each of the other project management functions, and Table 11.1 provides further examples of how some typical risk events (defined in Chapter 1II.F) can surface in any one of these functional areas.' In the groupings tabulated, risk items are associated with the headings in which the impact is most direct. For example, poor organization is an indirect risk (threat) to Quality, but is shown under Human Resources, which it affects directly. Note that "contractor" is used in the broadest sense of anyone, including an employee, who undertakes to perform any work or service at a price, salary or wage.

-

Figure i.. Integrating Rkk With Other Project Management Functions I1

D. Project Risk Management Definition
This leads to the definition of Project Risk Management as folloW S : ~

Project Risk Management is seen as a formal process whereby risks are systematically identified, assessed and provided for. In other words, this function involves a deliberate sequence of identification followed by mitigation. The latter calls for both assessment and response, which may include such defensive actions as risk avoidance or dejlection by allocating appropriate risks to insurers or by other contractual arrangements; a careful risk assessment or detailed impad analysis; response planning and contingency planning, such as the development of alternative workarounds (discussed in Chapter V1.B); and the provision and prudent management of a budgeted contingency allowance. Not only are the uncertainties in most projects numerous, but they may also be interrelated. This affects project results in complex ways. It tends to lead to under-estimation of risk which makes it difficult for management to be confident in identifying

Project Risk Management is the art and science of identifyzng, assessing and responding to project risk throughout the life of a project and in the best interests of its objectives.
Figures 111.1 and 111.2in Chapter 1 1show a Risk Management 1 breakdown structure foiiowing the typical hierarchy of PMI's Body of Knowledge Management Functions. That is to say the function itself is level 1, followed successively by processes, activities, and finaiiy techniques at level 4. As illustrated earlier, and as shown in the figures, the Risk Management processes are Risk identification; Risk Deflection (Insurable), Impact Risk Assessment; Response Planning; the Response System; and the Application of the resulting Data.

Chapter II

Integration, General Approach and Definition

Risk Management Variation of Risk Factors Through the Project Life Cycle As anyone who has been associated with any sizeable project weil knows, the life cycle of a project is very dynarnic, i.e., characterized by rapid change. It should not come as a surprise to learn that the project risk factors are also subject to considerablechange during the project life cycle. The nature of this project life cycle and some of its special charactenstics are discussed in detail in PMBOK Handbook Volume 1, A Frammork for Project and Program Management lntegration. For purposes of this handbook, it is worth noting that a typical project is made up of four generic phases, consisting of wncept, develupment,implementation,and tmination and that these in turn are broken down into sfages specific to the industry or area of project application. In addition, the first two generic phases constitute project planning, while the last two constitute project realization or accomplishment.The nature of this project time frarne and some of its special characteristics are diçcussed at greater length in PMBOK Handbook Volume 1, A Frammork for Projecf and Program Management lntegratiun, Chapters I1 and 1 1 1. The significance is that opportunity and risk generally remain relatively high during project planning but, because of the relatively low leve1 of investment to this point, the amount at stake rernains low. In contrast, during project accomplishment opportunity and risk progressively fall to lower levels as remaining unknowns are translated into knowns. At the same time, the amount at stake
Project Life Cycie I* -

Table 1 . . Typical Functional D i s t r i b u t i o n of Controllable R i s k Items 1 1

PM integration Risk Events
Incorrect start of integrated PM relative to project life cycle

Scope Risk Events
Changes in scope to meet project objectives, e.g., regulatory changes

Quaiity Risk Events
Performancefailure, or environmental impact

Risk Conditions
Inadequate planning, integration or resource allocation (Anything which reduces the probability of properly determining project objectives, i.e., anything which directly or indirectly reduces the probability of project success.) Inadequate, or lack of post-project review

Risk Conditions
Inadequacy of planning, or planning lead time Poor definition of scope breakdown, orwork packages Inconsistent, incomplete or unclear definition of quality requirements Inadequate scope control during implementation

Rlsk Conditions
Poor attitude to quality Substandard design/materials/ workmanship Inadequate quality assurance program

Time Risk Events
Specific delays, e.g., strikes, labor or material availability, extreme weather, rejection of work

cost Risk Events
Impacts of accidents, fire, theft Unpredictable price changes, e.g., due to supply shortages

Risk Risk Events
The risk of overlooking a risk Changes in the work necessary to achieve the scope

Risk Conditions
Errors in estimating time or resource availabilrty Poor allocation and management of float Scope of work changes without due allowance for time extensions/acceleration Early release of competitive product

Risk Conditions
Estimating errors, including estimating uncertainty Lack of investigation of predictable problems Inadequate productivrty, cost, change or contingency control Poor maintenance, security, purchasing, etc.

-

Risk Conditions
Ignoring risk or "assuming it away" Inappropriateor unclear assignment of responsibility/ risk to employees/contractors Poor insurance management Inappropriate or unclear contractual assignment of risk

/- o l * t Ta

/

*
Phase 1
CONCEPT

Plan

Accomplish
Phase 2

Conceive

/

DEYELOPMENT

Develop

I

+
Finish

Phase 3 Execute

IMPLEMENTATION

1

Phase 4
TERMINATION

Human Resources Risk Events
Contractor insolvency Claims settlement or litigation

Communications Risk Events
Inaction or wrong action due to incorrect information or communication failure Period when Highest R i s k are Incurred

Risk Events
Strikes. terminations, organizational breakdown

I I

/

Period of

\

Risk Conditlons
Unenforceable conditions/clauses Incompetent or financially unsound workers/contractors Adversarial relations Inappropriate or unclear contractual assignment of risk

Risk Conditions
Conflict not managed Poor organization, definition or allocation of responsibility, or otherwise absence of motivation Poor use of accountability Absence of leadership, or vacillating rnanagement style Consequences of ignoring or avoiding risk

Risk Conditions
Carelessness in planning or in communicating Improper handling of complexrty Ladt of adequate consuitation with project's "publics" (internavexternal) Figure 1 . . Typical L i f e Cycle Profiles Risk vs. A m o u n t at Stake 12 R.M. Wideman, Project Management Course, 1990

-

C hapter I I

Integration, General Approach and Definition
steadily rises as the necessary resources are progressively invested to complete the project. n These trends are shown graphically i Figure 11.2. The figure also shows that the period of highest vulnerability to risk occurs during the last two phases. At this time, adverse conditions rnay also be discovered as a result of acceptance testing and start-up of the project. The purpose of risk management must be to influence the project planning such that both uncertainty-risk and amountat-stake are reduced to acceptable levels throughout the project life cycle.

Risk Management
Phase Two Assessment Having identified the range of possible risks, the next step is to assess them. The purpose is to determine their ranking or status i terms of type, impact and probability. This may range from a n simple attempt at subjective evaluation to a more serious attempt at measurement. Due to their nature, or simply through lack of relevant data, however, it rnay be found that many of the risks defy direct measurement and a more in-depth impact analysis becomes necessary. Risk assessment typically involves input from all of project management functions, as shown in Figure 11.1. Consequently, a major benefit of risk management, particularly early on in the project, is the integrating and team-building effect experienced by members of the project team. Risk assessment methodology is discussed in Chapter TV and impact analysis is described in some detail in Appendix B. In addition, many sophisticated techniques have been developed in support of these processes, a number of which are briefly described in Appendix C. Phase Three Response Mitigating project risk requires, first, establishing an appropriate system strategy, then taking out insurance as appropriate against those risks that are insurable, and finally, planning specific actions to deal with the remainder. These rnay range from simple decisions to accept the risks as they are, especially on a small project, to a comprehensive plan for deployment of resources to control a risk event, should it occur, where the event rnay be far reaching (e.g., labor strife) or urgent (e.g., fire, accident). Phase Four Dowmentatlon Final documentation is a vital part of any project activity, though regretfully often overlooked. This is just as true of risk management. The purpose is to build a data base of reliable data for the continuing evaluation of nsk on the current project, as well as for improving the data base for all subsequent projects. Each of these phases are discussed in greater detail in the following chapters.

-

F. Four-Phase Approach
In its most simplistic form, project risk management consists essentially of four process phases as shown in Figure 11.3, narnely: Identification Assessment Response, and Docurnentation

-

RISK MANAGEMENT

I
IDWTIFICATION ASSESSMENT

r

RESPONSE

DOCUMENTATION

-

Figure 11.3. Risk Function Breakdown: Four Processes

Phase One Identification This phase consists of identifymg all the possible risks which rnay significantly impact the success of the project. Conceptually, these rnay range from high-impact/high-probability,through high-impact/low-probability, low-impact/high-probability to low-impact/low-probability.Obviously, the high and medium risks, including accumulations under any one item of risk, should receive the most attention. Moreover, combinations of risk which together pose a greater threat than each individually should not be overlooked. In order to identify all the potential risks to a particular project, it may be necessary to undertake a risk identification program. This might involve soliciting the considered opinions of knowledgeable persons associated with the project or similar projects, or conducting a "brain-storming" type of workshop amongst the project team. But always remember, "the alligators that you do not see, are the ones that bite you!"

-

1. C. Quaife, text contributed September, 1989. 2. PMBOK, 3/28/87, p E-2.

Risk Management

Chapter III Risk Identification
A. Risk in Corporate Business Management
In corporate business management, nsk is typically divided into two basic types. The first of these, Business Risk, includes the inherent chances of both profit or loss associated with the particular business endeavor. Business entities employ staffs of specially trained managers, professionals, technicians, and skilled workers in order to increase the chances of profit and reduce the chances of loss. The essential purpose is to maximize profits. The second type of nsk is usually called Pure, or Insurable Risk. Insurable risk differs from business risk in that it involves only a chance for loss and no chance for profit. For example, insurable risks can be further divided into four general categories relating to the chances of loss: direct property, indirect property, liability, and personnel. Obviously, direct property loss involves the destruction of property such as by fire, flood or wind storm. Indirect loss is somewhat more subtle and involves, for example, the extra expenses associated with renting alternative temporary accommodation or equipment following its damage or destruction, or the loss due to business interruption if operations cannot be continued because the replacement is not immediately available. Liabiiity loss, of course, involves the chance of a member of the public filing a lawsuit for bodily injury, personal injury, or property damage against the contractor. Finalíy, personal losses generally involve injuries to employees such as those contemplated by Worker's Compensation Laws.

B. Risk in Project Management
On most projects, responsibility for Projed Riskis so pervasive that it is rarely given sufficient central attention. Moreover, the reader is reminded that not all risk events are independent. Indeed, the total amount at stake on a project may be highly dependent upon a series of interacting events. The old adage "It never rains but it pours!" is not an uncornmon experiente. In addition, a series of nsk events can, and frequently do, cross traditional functional responsibility boundaries, which, with their classic difficulties of coordination and rapid response, can lead to disastrous consequences. Ultimate responsibility for identifying risks to a project and their subsequent treatment must rest with the project sponsor. The very threat posed to the successful achievement of project

Chapter III Risk Identification
objectives, as expressed in terms of scope, quaiity, time and cost, should be sufficient reason for the sponsor to recognize this responsibility. A particular case of risk is where the project leader must take steps to keep the project on schedule, but must act with insufficient information to make a sound decision. Such risks should be calculated, not reckless, and the absence of the requisite information clearly recognized. It rnay be that the collection of the information is either too costly, too time consuming, or simply unavailable. In such situations it is advisable to have a contingency plan prepared so that when the information or results do become available, and they prove to be negative (i-e.,bad news), there is a "fall-back position to tum to. In Research and Development projects, for example, unknowns that can t m into unpleasant surprises rnay include technicai solutions that do not work as expected; experirnents that fail; unanticipated by-products or side-effects, such as the undesirable side effects of a new dmg; being eclipsed by an unanticipated superior product announcement, or by an overriding patent application by a competitor; market research indicating lack of customer acceptance; escalating product development costs; or simply product reliability, quality or producibility difficulties. In R&D, some of these risks rnay be mitigated by conducting parallel development paths, with the view that if one approach does not work, then perhaps another will. This rnay cost more, but rnay be worth it in order to maintain flexibility, and so reduce the overall risk and elapsed development time.'

Risk Management
By their very nature projects are risky business. This interesting but simplistic approach is hardly adequate for project purposes, but it does underscore the fact that once all the risks seem to have been thought of there rnay still be a few remaining! Another approach is to c l a s s e risks according to their impact * on the project. For example: Scope risks - risks associated with changes of scope, or the subsequent need for "fixes" to achieve the required technical deliverables Quality risks - failure to complete tasks to the required leve1 of technical or quality performance Schedule risks - failure to complete tasks within the estimated time limits, or risks associated with dependency network logic Cost risks - failure to complete tasks within the estimated budget allowances Unfortunately, many identifiable risks will have an irnpact on two or more of these areas, particularly both schedule and cost, so that this leads to significant overlapping and potential double counting when it comes to making offsetting provisions. Yet another way of classifymg risks is to separate them according to their nature. For example, discrete one-time risk events such as fire and theft rnay be distinguished from those. that are time-scaled, such as with flooding or earthquakes, because in the latter case the probability and magnitude of occurrence varies with the period of time selected. Such risks are typically insurable, and corporate management usually draws a distinction between insurable risk and business risk, where business risk is risk arising from the business venture itself. Deliberately chosen risks, such as correctly identifymg project objectives for a venture opportunity, rnay be distinguished from those which are latent, i-e.,inherent in a situation or product, such as changes in market conditions ora faulty mechanical part. Again, for any particular project, some risks rnay be considered to be sufficiently remote or catastrophic as to be outside of the realm of project responsibility. Obvious examples include a change in political direction or the financia1 collapse of the sponsoring organization.

C. Types of Risk
Risks rnay be classified in a number of different ways. For example, one way is to describe uncertainties (and hence opportunities and risks) in terms of k n m s , known-unknowns, and unknm-unknowns. Aknown is an item or situation containing no uncertainty. An example of a known in our personal lives is death-it wiil happen and there is no uncertainty about it. Unknowns are those things which we know exist but do not know how they will affect us. A known-unknown is an identifiable uncertainty. An exarnple of a known-unknown is our electricity bill-we know that we shail get one next month but do not how much it wiil be. Another example is cancer. We know that cancer exists, but do not know if we shall fall victim to it. An unknown-unknown is simply an item or situation whose existence we cannot imagine. For example, before the first case was reported, AIDS was an unknown-unknown. Now, however, since we know that AIDS exists, it is a known-unknown like cancer. Obviously, there can be no example of an unknown-unknown since, by definition, its existence camot be imagined2 The spectrum of risk events will obviously vary between projects, but projects are launched to take advantage of opportunities and, as explained earlier, opportunity and risk go together.

D. Project Risk Identification
In order to deal systematically with the variety of risks encountered in project work, a more useful approach to risk identification is to class* the types of project risk according to primary source (rather than effect). This will also facilitate more effective management. The PMBOK categorizes sources of risk as follows: External, but unpredictable External predictable, but uncertain Interna1- non-technical Technical Legal

Chapter III Risk Identification
It wiU be seen that this form of classification also provides the opportunity to rank the various risk groupings according to ability to manage effectiveresponse (i.e., their relative controllability). The degree of ability to manage the response is, of course, independent of probability, amount at stake and, hence, risk event status, discussed in Chapter 11. This part of the risk management breakdown structure is shown in Figure 111.1, with the groupings generally arranged from low ability to high. More detailed listings within these typical groupings are given in Appendix A. The sequence of mitigation activities form the remainder of the risk management breakdown structure as shown in Figure 111.2. These activities are discussed in subsequent chapters.

Risk Management
Maximize reliability with rninimurn production downtime, as with a public utility; Maximize safety with minimum environmental impact, as with a public transportation service; or Maximize flexibility and use of interna1resources, as in the case of a commercial service to the public? Such questions must be established at a very early stage of the project, and the answers and resulting decisions themselves often must be made in a high degree of uncertainty requiring qualitative judgement. In any case, the results will have a considerable effect on the riçk characteristics of the project. These decisions, properly handled, typically involve multidisciplinary effort, and perhaps specialists who are required to examine the project in detail. In a significant and beneficial integrative risk management effort, information flows from one group to another in the form of most-likely estimates of project parameters. Gradually, an overall approach is built u p around these estimates, which may then be tested against other possible criteria, such as sensitivity to econornic conditions, including prices and demand; competition; life cycle econornics; other project alternatives; and so on. The technical configuration of the project may also be subject to design alternatives, in which appropriate questions would be: With respect to which project parameters is this design feature required? What is the likelihood of variation of the referenced parameters, and hence the need of the design feature as described?

E. Project Risk Configuration Identification of risks associated with a particular project commences with an understanding of the project itself. What is the project scope, i.e., the project deliverables, and indeed what are the underlying project objectives? Answers to these questions will have significant impact on the selection of probable risks to be considered on the project, and, in particular, will impact decisions on alternative project strategies and work-arounds for identified problems. For example, in selecting a system for project delivery, e.g., in-house resources or outside single or multiple contracts, is the underlying objective to:
Maximize return on investment, as in a commercial venture to supply goods; Minimize cost and financia1 risk, as in the case of a project undertaken by a non-profit organization;

TWO KINDS: DISCRETE EVENTS )) COMINUOUS SCALE

I

RISK IDENTIFICATION

*

qulMANAGEMENT

MANAGEMENT

I

//-

I

EXTERNAL UNPREDICTABLE

I I
-

EXTERNAL PREDICTABLE
Market Risks Operational Environmental ImpactF Social lmpacts Currency Changes lnflation Taxatior.

I

(

INTERNAL NON-TECHNICAL
Management

I I

TECHMCAL

I I

LEGAL

I

.

Regulatory Natural Hazards Postulated Events Side Effects Cumpletion

Sdiedule
Cost Cash Flow Loss of Potential

Changes in Technology Performance Risks Specific to Technobgy Design Sheer Size or Complexity of Project

Linces Patent Rights Contractual Outsider Suit Insider Suit Force Majeure

pl u r ASSESSMENT RESPONSE Baseline 8 Structurina Screening Quantiication Mcdeling Overal Assessment Report Findings STANDARDS ALTERNATIVES
I
I I

Historical Database Cunent Project

Database
Post Project Assessment and Archii üpdate

I

Definitions Policies~Procedures Responsibility Alblion Monitor 8 Review Risk Model Update System Adjustments

Direct Property Darnage Indired Consequential Lws Legal Liability Personnel Related

-

Avodance Absorbtion Deflection Contingent Planning Unforeseen Allowance

Figure iI1 Risk Managemenk Identification I..

i.. Figure Ii2 Risk Managemenk Mitigation

Chapter III Risk Identification
What is the likelihood that this design feature will perform as expected? What are the implications of this feature on project parameters and objectives which constitute project success? From these questions, the costs, benefits and risks of a particular design feature can be evaluated under various scenanos and the relative benefit determined. None of these questions can be answered properly, however, unless they have been dealt with early in the project planning and evaluation phase, and explicit estimates made of the implications.

Risk Management

Chapter IV Risk Assessment Goals and Methodology
A. Risk Assessment Goals The goals of risk assessment are to:
Increase the understanding of the project in general. Identify the altematives available in delivery and methods. Ensure that uncertainties and risks are adequately considered in a structured and systematic way, which allows them to be incorporated into the planning and project development process. Through direct examination of these uncertainties and risks, establish the implications of these on a11 other aspects of the project. Major benefits of these goals include: Greater information is made available during the course of project planning and decision making; for example, estimates of uncertainty of project performance and viability should evolve; The project objectives themselves may be called into question, and hence improved upon; Improved communication between members of the project team, and other project stakeholders, where appropriate; Confidente that the true implications of uncertainties and risk have been exarnined and incorporated into the project plans; Documented support for the project contingency allowance, and a basis for its application management; Reduced probability that the realization of the project will be sub-optimal, either by identifying weaknesses or by forcing improvements during the project planning phases; Consequently, a reduced likelihood of dismptive changes during implementation; Hence, substantially increased chances of project success. Properly handled, the risk assessment process should promote the language of probability and the use of its associated mathematics in risk analysis. It is best handled by examining the individual elements of the project in some detail and determining their relationships, because most people are more able to

F. Risk Factors
All project risks are characterized by the following three risk factors?

- precisely what might happen to the detriment of the project; 2. Risk probability - how likely the event is to occur; and 3. Amount at stake - the severity of the consequences.
1. Risk event

With this data, the risk ment status (criterion value or ranking) of a given risk event can be detennined by the following relationship:
Risk Event Status = Risk Probability x Amount at Stake

Some risk events are characterized by low probability and high severity, while others are the reverse. Clearly, the most serious nsks are those involving both high probability and high severity. As noted earlier, many risk events cannot be treated as simply discrete and independent as the total amount at stake may increase substantially as a result of a series of interacting events. Such a situation cails for careful examination and special analytical techniques.

1. W.N. Hosley, All-Tech Project Management Services, in letter dated April19,1990. 2. M.W. Curran, Decision Sciences Corporation, in letter dated November 26,1990. 3. D.C. Fraser, Risk Minimisation in Giant Projects, International Conferente on the Successfd Accomplishment of Giant Projects, London, England, May 1978.

Chapter IV Risk Assessment Goals and Methodology comprehend the parts than the whole. In a formal quantitative risk assessment, i.e., a risk impact analysis, a mathematical means is developed to integrate the complex relationships between the detailed elements of available information and the information possessed by numerous experts while preserving the experts' uncertainty of 0~ini0n.l Bear i rnind that the objective of project management i always n s to achieve project success through "participant satisfaction." Since the long-tem viability of the resulting product is often a key element of success, it may also be necessary to conduct a comprehensive product life cycle risk analysis duríng the conceptual phase of the project. This would be particüíarly tme if the projed is large and/or complex, the product has an extended life as in buílding and engineering works, or i.part of an extended program of projects. In this case the project risk assessment, which essentially covers the functions of technical scope, quality, time and cost, may form part of this more comprehensive product risk analysis. 8. Assessment Methodology Selection Perhaps the greatest impediment to the acceptance of risk management as a normal part of project management is the realization that project risks are so many and various that more than superficial traditional attention might thwart the project at the outset. This attitude denies the contribution that closer examination of some of the risks to a particular project can bestow on its ultirnate success. The issue is: Having identified possibly a large nurnber of risks, which should receive attention? In fact, no risks should be entirely ignored, but many of the lesser risks can be provided for by the conventional contingency ailowance approach. Clearly, the risks that should receive the closest attention are those that could have both the greatest impact on the project as well as the most likely probability of occurrence. To some extent, the selection of these particular risks is an iterative process in which prelirninary analysis may indicate the need for further study. Figure IV.l shows a conceptual flow diagram in which risk events are first categorized and then assessed for severity and probability in order to arrive at a criterion value on which a priority ranking can be based. The extent of the assessment at this stage should be governed by the project's risk management policies,,but in any case, the categorization should be closely aligned to the project's work breakdown structure. Consequently, the possibiiity of a significant impact as a result of some combination of apparently minor events must not be overlooked. For example, a succession of relatively insignificant schedule delays, perhaps as a consequence of a spate of untimely scope changes, could become highly significant. The effectcould be to completely miss a "window of opportunity" such as reaching a market before a competitor, forestalling technological obsolescence, or construction in the summer season rather than in winter.

Risk Management

PROJECT RISKS

RECURRING CONDITIONS

NON-RECURRING EVENTS

OBJECTIVE ANALYSIS

SUBJECTIVE ANALYSIS
H'GH

l

2

PROBABILITY OF OCCURRENCE

-t

O B

5 (PRIORITY) j
..%:9:..:>=r:r=~:,%~:..:~:..-'

HIGH RISK

i

L

I
SEVERIN OF CONSEQUENCES

RISK CRITERION VALUES (RISK EVENT STATUS)
Figure IV.1. Comparing Severity of Project Risks
R.M. Wideman, Project Management Course, 1991

Simple Assessment Development A simple risk assessment may be conducted by stepping through the sequence shown in Figure IV.2. As a prelude to a better understanding of the relative sigmficance of the findings, however, a Risk Baseline should be established based on the organization's externa1 "status quo." T i will establish the risks hs to the sponsoring organization associated with not carrying out the project at ail. The internal intrinsic worth of the project should also be documented.
Step 1 Select the risk events, or series of related events, to be examined. Prioritize these for attention according to the initial selection discussed earlier. Certain risk events may be screened out from further consideration, as a matter of policy, even though the event could have a major impact. For example, an "out-of-scope" risk event might be one that seriously impacts other activities of the sponsoring organization, or is common to more than one project. i this case appropriate ren sponse pianning might be incorporated into the organization's business planning, rather than at the project level, to avoid overlap and duplication. Blanket corporate insurance coverage is a prime example.

Chapter IV Risk Assessment Goals and Methodology

Risk Management Note that amount at stake and criticality may vary with time, i.e., according to the stage in the project life cycle, as discussed earlier in Chapter 1I.E. In most cases, the amount at stake and criticality can be amved at by a simple examination of the available data and some subjective judgment. In complex situations, however, it may be necessary to develop some form of mathematical model and conduct a series of computer m s .

PROCESS
-

I D E

N

A N A

T I
F

Y
STEP

L Y S E

R E S P O N D

D O

C
U M

E N T

EVENT

PROBABILITY

Step 4 Having identified the consequences and their significance, this step involves planning to mitigate the likelihood of the risk event(s) in question, and/or developing suitable responses and contingency plans, as discussed ,in Chapter VI. It may even be necessary to gain more insight and gather additional information to complete this step. Either way, it should be the most creative step of all because it provides the occasion for converting risks into opportunities. Step 5 The final step in the process is to accumulate the results of the assessment in a set of "Conclusions and Recommendations" such that appropriate management decisions can be made with full knowledge of the apparent risks involved. Either the residual risks must be accepted, or the project abandoned. By following these steps the management of risk and uncertainty can be directly incorporated into the early project planning process as well as dealt with expeditiously during the course of project execution.
Risk Quantlfication

CONSEQUENCE

Figure IV.2. Impad Analysis Matrix Sequence

Step 2 Assess the probability associated with the risk event(s1. This is perhaps one of the more subjective steps, although there are a number of procedures which can help. An estimate of the degree of uncertainty may be arrived at by:
Influence diagrams Risk contribution analysis Probability distribution Probability trees Riskmodelling Sensitivity profiles Where the determination of probability is particularly elusive, but important, there are more elaborate techniques available as described in Appendix C . Beware, however, of overconfidence in the accuracy of the results of these various approaches. At best, they are estimates based on good experience and thoughtful opinion.

Step 3 Assess the consequences and severity of the risk event(s) by determining:
the amount at stake, and the criticality.

The application of the various techniquesnoted in Steps 2 and 3 above can provide insight into risk event interdependencies,the merits of further detailed consideration of specific risks, and the manner in which combined effects of risk events might be modelled mathematicaliy. In such an analysis, especially on large projects, it is often necessary to develop a further breakdown in which each activity is numbered and docurnented for reference. Using this breakdown, the risks within each activity are identified by mentally stepping through all aspects of the activity to produce a comprehensive list of uncertainties. As with the project work breakdown structure, this breakd o m serves to focusdiçcussion,to aid in identification of all risks, and to provide a basis for formalizing dependency links within the project. In this way a model may be developed in which the variables are represented by discrete probability distributions having specified linkages. This allows maximum flexibility in representing distribution shapes as well as offering mathematical simplicity. It also paves the way for solvingcomplex combinations of dependent and independent variables by repetitive computerized calculations.

Chapter IV Risk Assessment Goals and Methodology
Where risk combination is analyzed by such modeliing, three levels of model are typically required. These are:
I. For detailed analysis of the joint impact of a small nurnber of

Risk Management
Sirnple Probability

risks within an activity,
2. For examining the joint effects of all risks within an activity, and

3. For examining the broad overall impact of risks from severa1 or

all activities. This can be conceptualized as the successive summarization of a large probability tree and the resulting output shows overall distributions as they impact cost, schedule and quality. These diçtibutions can be displayed graphicaily so as to show the relative importance of each contributing risk, as well as their cumulative effect. Such risk analysis is diçcussed in more detail in Appendix B. In-depth project risk impact analyses are generally the purview of specialists in risk analysis who are familiar with the vanous technical aspects of the project management application in question. This may require a significant cornrnitment of time and resources and may only be appropriate where there is substantial uncertainty, the stakes are high, and there is a need for significant management attention.

To provide a better understanding of simple probability, consider the following question and answer: "What is the probability that we shall get approval for our project next month?" "It looks good, say, about 75%!"So what is the estimated probability of this event occurring? 75%?However, it alço means that there is a 25% probabiiity that approval will not be obtained. Note that the probability of the event occurring Pr(Event) plus the probability of the event not happening Pr(No Event) equals one-always. What if there are two related events? For example, consider the following question and answer: "What is the probability that we shall have the scope defined by next month and that we shali get approval?" "Well, it still looks pretty good, say, about 80%that it wiil be ready and 70% that we shall get approval." If the two events are possible but not certain, then how likely is it that they will both happen?

But Pr(Event #I) is 80%,and Pr(Event #2) is 70%, so how likely is it that both will happen?

C. Advantages of Assessment Methodology
From the foregoing it can be seen that there are additional benefits which derive from this assessment methodology by providing:
1. The vehicle for incorporating uncertainties directly into the

That's barely over a 50-50 chance. Suppose that only one of these things is necessary before starting project planning. What is the probability that we shall start project planning? Pr(No Scope) x Pr(No Approval) = Pr(No Planning) 0.30. x. 0.20 0.06 = 6% The probability that neither will happen is very low, so it is 94% likely that we will start planning. Another way to look at this problem is in three parts: Pr(Scope) x Pr(Approva1) = 0.70 x 0.80 = 0.56 Pr(Scope) x Pr(No Approval) = 0.70 x 0.20 = 0.14 Pr(No Scope) x Pr(Approval) = 0.30 x 0.80 = 0 2 .4 0.94 or 94% likely
More Complex Probabiliiy Probabiiity ranges are more complex to deal with, especially in project work. For example, if a given human activity is repeated many times, ostensibly under identical conditions, then the actual durations experienced will nevertheless not be identical. This variation will be due to a number of influences impacting the activity such as human productivity. Theoretically, if the frequency of occurrence (i.e., the number of times that a particular duration occurs) is plotted against the time taken for the activity, the resulting plot wiil produce a "Gausian" distribution curve,

2.
3.

4.

5.

6. 7. 8.

project management process of planning, development and implementation of the project A clear understanding of the overall project's goals, objectives, scope definition and feasibiiity What the risks really are, which are the most significant, and hence which should receive attention to lead to the most risk reduction The models and techniques by which the variability and uncertainty of estimates can be conveyed quantitatively An information base of quantitative and order-of-magnitude data to support trade-off decisions, such as choices between cost and performance, or the comparison of different options Amore rational basis for contingency planning and evaluation A more consistent and workable project plan An early warning for risk

It is better to avoid risks now than to encounter them later.

D. The Basics of Probability
Probabiiity may apply to simple on/off or go/no-go type situations such as getting approval or not getting approval, or it may be more complex and apply to ranges of probability as encountered in estimating time and cost2

Chapter IV Risk Assessment Goals and Methodology
popularly known as a bell curve. The bell curve is typicaily symmetrical about its highest frequency value, in which case it is described as a normal distribution. The probability of any particular time being taken is, strictly speaking, its number of occurrences divided by the total number of times the activity was repeated in the whole sample. This fraction may be expressed as a percentage. For example, the probability (chances) of landing a "heads" or "tails" in a coin toss is 0.5 or 50%.Similarly, the chances of puíling any given playing card from a full deck is 1 in 52, or approximately 2%. In project work, two practical difficulties arise with the application of this theory. In the first place, a set of observations rarely exists upon which a discrete probability calculation can be made, and rarely is there the opportunity to carry out repeated runs of an activity during project planning in order to make the calculation. Consequently, where future events are being postulated, it i s necessary to rely on speculation. This leads to the second difficulty. When people are asked to speculate on probability, there is typically a tendency to be optimistic. This may be due to natural human optirnism, but is more likely due to it being easier to overlook obstacles than it is to account for them. Consequently, such bell curves of probability are rarely symrnetrical. Two examples are shown in Figure V.1, Chapter V. These probability distribution curves show the many values that an element rnight take. The concept is used in Range Estimating (see Chapter V.C). When speculating on the probability of future events, it is usual to establish three values in order to fix the shape of the curve. These values are the two outer limits of the element plus the value which has the highest probability of occurrence, i.e. the "most likely." This simpler approach is used in PERT calculations (see Chapter V.B). Two examples of how these rnight be expressed: The cost of project planning will fali between $x and $y with the cost distributed "normally" around $2; Activity #B116 has a low value of "o" days, a high value of "p" days, a most likely value of "m" days with a triangular (square, stepped, bell, etc.) distribution. The "mean" of a probability distribution curve (i-e.,the value at which there is 50% of the total area under the curve on each side) is known as its "expected value," and this expected value is found by taking: (the value an element can take) x (probability that it will take that value)

Risk Management The "most likely" value referred to earlier is that value which has the most likelihood of occurring. It is only the same as the "expected value" if the distnbution is symmetncal around the "most likely." Note that the sum of the "means" (expected values) is the mean of the sums (total).That is, if the total cost of a project = the sum of WBS items#l-100, then the "expected"total cost is the same as the sum of the "expected"costs for each WBS item #1-100 since these are all amved at by calculation from the given observations. The sum of the separate "most likely" values, on the other hand, is not necessarily the "most likely" for the whole project. When ail is said and done, the project manager should be wary of falçe impressions of accuracy generated by extensive calculations. The assessment of the probability of an event occurring is only as good as the available historic data upon which the assessment is based, or the quality of the experience and opinions of those making the assessment.

E. The Quality Risk
The goals of risk management are to increase understanding of the project, hence improve project plans, system delivery selection, and especiaily to identify where the greatest risks are likely to occur during the phases of project accomplishment.This helps to estabiish where management can best focus its attention during the project and much of that attention will be concentrated on containing potential overruns of schedule and cost. Presuming that the project is not complete until the entire scope is accomplished, there wili nevertheless still remain a major area of project risk. This risk can best be expressed by the question: "What if the project fails to perform as expected during its operational life?" This may weil be the result of less than satisfactory quality upon project completion, and is especially true if quality is not given due attention during the project life cycle. Since the in-service life of the resulting product is typically much longer than the period required to plan and produce that product, any quality shortcomings and their effects may surface over a prolonged penod of time. Consequently of aU the project objectives, conformance to quality requirements is the one most remembered long after cost and schedule performance have faded into the past. It follows that quality management can have the most impact on the long-tem actual or perceived success of the project. This may be demonstrated by considering the long-term cash flow, including project costs, of a commercial venture as displayed in Figure N.3. As the figure shows, the intended return-on-investment could be thwarted by "poor" quality. Quality risk impacts may remain hidden or ignored, but are not forgiven if the project fails to deliver its long-term objectives.

s then summing the results, i-e., the expected value i the "weighted average" (possible values weighted by their likelihood of occurrence).

F. The Schedule Risk
It is possible to manage the "critical p a t h of a schedule activity network but not manage the project duration. This is

Chapter IV Risk Assessment Goals and Methodolsgy
DESIRED or INTENDED RETURN ON INVESTMENT

Risk Management

PROJECT

+
PRODUCT or FACILITY CASH FLOW SS (accumulated and discounied)

I

' PROJECT CONCEIVE EXECUTE OEVELCP FINISH

I

PRODUCTION

I
PRODUCT or FACILITY LIFE-CYCLE

I

Figure iV.3. The Quality Risk: Difference Between Success and Failure

because the schedule risk3 is the "highest risk path'' that contributes the most risk to project completion, and this path is not necessarily the critical path as determined by simple network analysis. In fact, the "likelihood of finishing on time" requires examination of the risks associated with aii the activitiesnecessary to reach completion. There is risk in the duration of every activity because any duration in the future is uncertain. Therefore, duration must be measured as a range, and this is typicaíly expressed in t e m of the low, most likely (or altematively, expected) and high durations associated with specified degrees of certainty. The extent to which the high-risk durations impact project completion wiiI depend on the logical relationships between activities and the skiilful management of available float. Note that the longest activities are not necessarily the "riskiest" (a long-duration activity could be quite reliable). Indeed, any activity may be "highly risky,"and any such activity could delay the project, whether or not it is on the critica1path. Therefore, it is necessary to idenhfy and rnanage aii the activities that could contribute to the most delay to the project, which are not necessarily those on the critica1 path as observed earlier. This suggests that where significant project activity risks are involved a standard "critica1path method" (CPM) may be of only limited vaíue. However, the CPM approach could suffice if the "expected" (calculated) durations are used rather than the "most Iikely" (see Section D above for discussion of "expected" and "most likely"). After considering significant activity duration risks on a particular project, it is quite possible that a sound management strategy would be to forego the "expected" completion date by severa1 days in order to reduce the overail project risk. To make this determination it would be necessary to combine activity risks along alternative paths of the schedule network. On sirnple networks this is relatively easy but for complex networks the process is not seu-evident nor intuitive.

Figure N.4. Simple Example of Network Completion Risk
(MacKrimmon and Ryavec in Archibald & Villona) Hulen, PMP Ceriification Workshop - Risk M a n a g e m e n t ,

D.T.

PMI Orange County Chapter, 1991, p 5 2 .

Consider the activity network shown in Figure IV.4, having low, most likely and high durations as shown in Table IV.1, Network Activity Summary. Calculated means assuming a symmetrical distribution are shown in the fifth column of the table. Table IV.1. Network Activity Summary
ACTIVITY A-B B-C C-E B-E A-D D-E LOW
8 4 O 1 4 1

MOST LIKELY 9 5 O 6 9 2

HIGH
1O 6 O 7 14 7

MEAN EXPECTED
9 5 O 4.7 9 3.3

Three ways of evaluating duration along alternative network paths are shown in Table IV.2, Network Path Evaluation. The question is, which path is the most risky? The exercise gves multiple answers and, hence, mixed signals. Table IV.2. Network Path Evaluation
PATH A-B-C-E A-B-E A-D-E MOST RISKY
+

SUMOF MOST LIKELY
14 15 11 A-B-E

SUM OF MEANS
14 13.7 12.3 A-B-C-E

SUMOF HIGHS
16 17 21 A-D-E

The example is used simply to demonstrate that in the face of considerable uncertainty more sophisticated calculations

Chapter IV Risk Assessment Goals and Methodology
are needed in which degrees of uncertainty can be specified, probabiiity distibutions assigned, and relative activity risks assessed. Computerized mathematical models can be constructed to deal with these variables, but software is also now avaiiable off-the-shelf to assist the project manager by creating a number of probability projections expressed as " S curves (see Handbook Volume 1,Chapter V1I.F). The collective result to any given degree of certainty is then represented by the envelope of ovemding values, i-e., those generated by the nskiest near-critica1 activities.

Risk Management

Chapter V Computer Applications
A. Data Storage, Retrieval and Computation The use of computers, and microcomputers i particular, has n gained ground rapidly in recent years due to the v e r =rapid increase in the power of micro hardware. This has stimulated the development of elegant software capable of storing data in flat readily-accessibledata bases, or rapidly executing complex calculations which can produce robust graphical output. All of this greatly facilitates communication, an essential ingredient of successful project management. The capacity of the computer simply to store information for subsequent retneval is invaluable. Most projects involve large amounts of data which once associated with it become unique to that project. If data can be collected and subsequently readily searched and distributed to meet the progressive information requirernents on the evolving project, this in itself can go a long way towards avoiding the risks of mistakes, misunderstandings and misdirections. Data of this sort typically includes agreements, instructions, specifications, drawings, latest changes, progress coordination, current pnorities, and so on. Powerfui and sophisticated software is now available on the desk-top to support complex scheduling, range estimating and costing, spread sheet type calculations, statisticalquality analysis and, more recently, the possibility of artificid intelligence applications (see Chapter X-C). The computer helps mernbers of the project team to reduce the drudgery and time involved in handling large amounts of data and wearisome calculations. It also helps rapid retrieval of vital information for decision making. In addition, personal computers are establishing an increasing role in the transfer of electronic data and, together with the related "fax" machine, in the speed and reliability of project comrnunications. B. PERT and the Probabilistic Model One of the earlier applications of the computer was to the PERT schedule network analysis technique. PERT stands for Program and Evaluation Review Technique, a technique which was originalIy developed for complex projects with an innovative content and a high degree of uncertainty. Today, the t e m PERT Chart tends to be misapplied i . some software descriptions to refer to a project schedule when presented in the form of a logic network This is to distinguish it from a Gantt Chart, which

1. After D.T. Hulett, PMP CertificationWorkshop - Risk Management, PMI Orange County Chapter, 1991, p19. 2. Ibid., p3. 3. Ibid., p46-53.

Chapter V Computer Applications
presents the same data but in the form of a bar chart, i.e., horizontal bars representing activities plotted against time. The real significance of the PERT technique is that it recognizes that estimates of activity time duratioqs are just that-estimates. As was shown in Chapter I.C, opportunity and risk are closely allied and in a probabilistic world a "most likely" estimate has by definition an q u a l value set of opportunities and risks that define the "best case" and "worst case." Some risks will inevitably come true as will some opportunities sol as noted earlier, projed management must constantly pursue the latter to offset the former? This concept is reflected in the PERT technique by establishing for each activity in the network three time estimates which represent: The most optimistic time possible for the activity, o, Its most likely time, m, and The most pessimistic time, p.
If these three estimates are represented by time durations of o, m, and p, respectively, PERT allocates four times the weight to the most likely so that the expected time is then given by:

Risk Management
An assessment is first made of the maximum tolerable vanation in an estimate's bottom-line total cost which might result from a variation in any single element of that estimate. This threshold value is calied the critical variance. The cntical elements in the estimate are then identified as those whose actual values could vary from the target by such a magnitude that the bottom-line cost of the project would change by an amount greater than the critica1 variance. Ata given level of probability, the variations in the actual values will be either favorable or unfavorable and will range from highest to lowest. The deciding factor is not necessarily the magnitude of the element itself but its potential for variation. Consequently, the number of such elements in a typical estimate is not very large. Nevertheless, because of the large number of possible combinations, the number of potential outcomes in the actual bottom-lhe cost is indeed very large. Using a computer, these individual n uncertainties are put together i such a way that the uncertainty in the bottom-line total can be measured. The range of the resulting estimate is a simple but effective measure of its uncertainty. Surprisingly,the entire process requires only a modest amount of effort. Figure V.1 shows the probability distributions of two projects both of which have the same 50 percent probability of being completed for the expected value of $165,000. However, because of their respective distributions, project A is more likely to be completed close to the expected value than project B because the latter is less certain. Project B will require a higher contingency amount to achieve the same level of confidence in the estirnate as project A.
EXPECTEDVALUE

Expected Time =
6 A more complete dismsion describing the application of the PERT methodology to cost and schedule analysis, with simple worked examples, is provided in Appendix D. Yet, this technique gives rise to a fair amount of calculation, even for a relatively smali network, so it is ideally suited to the computer. However, just calculating the uncertainties of activity durations (and costs) may not be sufficient. The consequences need to be worked through the project network logic. There are a number of software packages on the market that use network simulation using probabilistic evaluation of success-dependent logic to amve at possible results, or to evaluate various alternative "what if' scenarios involving uncertainties. Such tools help to give insight into the sensitivities of the variables for puposes of project planning.

I
PROJECT A

C. Range Estimating
The application of probabilistic modelin to cost estimates is employed in the Range Estimating approach,' which is used as an adjunct to traditional estimating and not a substitute for it. The methodology relies heavily upon the application of Pareto's Law, Monte Carlo simulation and heuristics to idenhfy:

PROJECT

B

5

The mathematical probability that a cost overrun will occur The amount of financia1 exposure (how bad it can get) Risks and opportunities ranked in order of bottom line importance The contingency required for a given level of confidence

COST ESTIMATE
Figure V.1. Probability Distributions:Two Projects With the Same Expeded Values of Total Costs
After J.R. Adarns and M.D. Martin, A Practical Approach to the Assessment of Project Uncertainty. PMI Serninar/Symposium, Toronto. 1982. plV-F.6

Chapter V Computer Applications

Risk Management
and five damage scenarios, at three criterion levels, would involve 375 calculations for the one type of risk alone. Figure V.1. shows alternative formats for presenting such probability analysis results graphically. Figure V.2 illustrates curnulative risks plotted by source category to show how the criterion values vary with degree of certainty.

D. Risk Analysis
Risk analysis, especially as desaibed in Appendix B, is particularly suited to computerization because it permits large arnounts of data to be both stored and manipulated. Data is usualiy entered interactively in response to question/prompts at each stage, and such that it can be rapidly changed to suit different requirements, or "what if' enquiries. The computer also provides the automatic ability to present the results graphically, instead of through seemingly endless streams of hard-to-comprehend printouts. Data to be handled rnight include: Different project configurations Different damage/response scenarios and consequences for each configuration Associated risk probabilities Data for translation to equivalent doliar costs or time Calculations of probabiiity distributions, either in terms of some natural value (e.g., lost time or materiais), project dollar costs, or project schedule delay. Since the amount of data produced by these calculations is typically enorrnous, a variety of plotted output formats greatly simplifies presentation. For example, in Figure B.1 (Appendix B) a set of histograms shows the probabiiity level identified for each risk event magnitude. Figure B.2 shows the probabiiity of damage level if the risk event occurs, and Figure B.3 shows the probability of the criterion value being realized if the particular damage level is realized. For the case iílustrated, calculating the effects of a single risk in each of, say, five configurations, at five risk levels,

E. Knowledge-Based Risk Management In our post-industrial information-intensive age, corporate planning and decision making appear to be dominated by highpowered information technology based on worldwide database retrieval networks and powerful computers. Through such networks and databases, alrnost a11 requued data can readily be obtained, except, that is, knowledge that domain specialists have obtained through experience. Recently,artificialintelligence (AI),or its sub-specialty, experf systems (also called knowledge-based systems) has attracted special interest in order to deal with such domain specialists' knowledge. The expert system provides an integrated approach which facilitates risk identification by generating a list of the most significant uncertainty factors and their descriptions. This list includes the principal risks of all major parties involved in a project, including the uncertainty factors that affect productivity, cost, schedule, quaiity, and performance. The user seleds factors related

POSTULATED EVENTS RISKS IMPLEMENTATION I MAINTENANCE RISKS

\

TECHNICAL RISKS

NATURAL EVENTS RISKS

1

I

C CRITERION VALUE
CUMULATIVE DISTRIBUTION
Any point on the a m e indicates lhe pobabiiity P lhat lhe uilerion value CwillnaIbeexceeded

c -

CRITERION VALUE

C

CRITERION "ALUE

__I_)

- -

DENSITY FORM Any point on the m e indiites the pobabitii P lhat a p m uitefion value C wiU be inaared

REVERSE CUMULATIVE
Any point on me curve indicates the pobability P ttiat the mterion value C U beexceeded

I

I

1

1

I

I

I

I

I

1

.95

.9 6

.9 7 .98 PROBABILITY

.9 9

1 .O

Figure V.2. Graphical Presentation of Analysis Results
Aíter A.B. Cammaert c. 1986

Figure V3 Presenting Results of Analysis: Cumulative Risks vs. Criterion Value .. - Construction Project

After A.B. Cammaert c. 1986

Chapter V Computer Applications
to a given project from the list. Each of the general factors is further divided into subelements which provide the user with added detail. After identmng the uncertainty factors, the ex ert s stem goes on to ask questions about nsk policy, and so on.IP Figure V.4. shows an example of the breakdown stmcture of an expert system inference net for constmction risk management.

Risk Management
Application of expert systems in risk management can provide a practical model which not only considers traditional models, but alço expert knowledge, d e s of thumb, and professional expenences. An expert system provides information which is necessary for management to make decisions under the condition of uncertainty. In his book, Knowledge-Based Risk ~ a n a ~ e m e n t , " Kiyoshi Niwa develops the concept of a common expert system, then identifies the weak points in terms of ill-structured management domains. He then proposes a new concept of human-computer cooperative system to overcome these weakpoints. The key element is incorporating human intuitive ability into a computer system to improve its flexibility and applicability.Specifically, the system includes: A knowledge base A computer inference function Human intuitive ability A human-computer cooperative system to associate computergenerated inference and non-logical human intuition The purpose of the system is to facilitate the calculation of risks and improved management responses in the execution of projects by commanding more comprehensive data.

7
TOTAL
RISK

Labor

J I

b T l hy1
I
I
Delay Disputes

Envirorunenra 1

4

unce;ram;y :n Labor

Producr:viry

i I
----

I

Cncercalnry :n Ecuipnenr

h l" '

=t
Orders Design

Saf e r y

Figure V.4. Breakdown Stmcture of Identified Risks
After R. Kangari and L.T. Boyer. Risk Managernent by Expert Systerns, PMJ. March 1989, p41

1. P. Buckley, in a written response to the first draft of this handbook 2. M.W. Curran, Range Estirnating: Measuring Uncertainty and Reasoning with Risk, Cost Engineering, Vol. 31, No. 3, March 1989, p18-26, and in a letter dated November 26, 1990. 3. R. Kangari and L.T. Boyer, Risk Management by Expert Systems, PMJ, March 1989, p40-48. 4. K. Niwa, Knowledge-Based Risk Managemmt in Engineering, John Wiley &Sons, New York, 1989, p ix.

Risk Management

Chapter V I
A. Response Options

Risk Response and Documentation
The structure of project risk response was shown diagrammatically in Figure 111.2, Chapter 111. Conceptually, responses to project risks identified and evaluated as a result of the steps described in previous chapters may take any of the following forms. A nsk may be: Unrecognized, unmanaged or ignored (by default) Recognized but no action taken (absorbedas amatter of policy) Avoided (by taking appropriate steps) Reduced (by an alternative approach) Shared (with others, e-g., by joint venture) Transferred (to others through contract or insurance) Retained and absorbed (by prudent allowances) Handled by a combination of the above Certainly the choice will depend on the project, the risks and the circumstances, but the selection should be based on a clearly defined set of standards.

B. System Standards
The next step, then, is to set policies, procedures, goals and responsibility standards for risk management on the project in question. Where appropriate, nsk policy should be based on the principie that responsibility should be placed on the shoulders of n those who represent the source of the risk i question. This will establish the scope and framework for the risk management function, whether it is simply a recognition of a task to be undertaken by the project manager, or the responsibility of a specialist or team under his direction. Bear in mind that risk events will affect the project's cost, schedule, or quality of the work to an extent which depends on s the event and how it i handled. The overall project risk will also vary considerably through the project life cycle, as described in Chapter 1I.E. Potential impacts wiil increase as tasks with nsk events of high probability are undertaken and then decrease as the b u k of the work is completed. The project risk may also change substantially as a result of changes in the scope of the project or changes in the method of working. Consequently, continuous review of the situation, with appropnate adjushnents, is strongly recornmended.

Chapter VI

Risk Response and Documentation
Generally speaking, project risk policies and procedures should encourage the following preferred actions as appropriate: Develop an environment and/or work-arounds designed to avoid or minirnize selected risk events Reduce impacts by defensive planning Transfer the effect to others having control over the risk source Make appropriate allowance for retained residual risks Provide some flexibility in the contingency allowance procedures to permit some response to the unexpected The t e m work-around may need some explanation. A workaround is an altemative solution to a potential problem-"We can get around that by doing this." For exarnple, Requirements: Provide airtight food container. Solution: Sheet metal box with tightfitting lid. Risk: Nonavailability. Work-around: Substitute durable molded plnstic container wifh self-sealing top.

Risk Management

D. Response Planning
As a result of the methodology outlined above, a picture of project risk wili emerge. This will include where, when, and to what extent exposure may be anticipated. With this picture in hand it is then possible to formulate suitable risk management strategies, whether by way of absorption, adjustment, deflection or systematic contingency planning. Adjustment may simply involve the proper recognition of certain risks by appropriate modifications to the project's scope, budget, schedule, and quaiity specification, or all four combined. Adequate contingency allowance and good control, even on a tight budget, will reduce the chance of overrun. A logically developed schedule with attention to resource requirements and conflicts will reduce the probability of schedule ovemn. But in the course of the enthusiastic initiation of project implementation, how often has one heard the battle cry "h objective is to buiid the best there is!" Not surprisingly, such an unrealistic definition of quaiity results in a very high risk indeed of the project being unsuccessfd due to cost and time overruns as well as possible failure to achieve even more realistic performance objectives! Deflection involves the transfer of risk by such means as: Contracting out to another party Insurance or bonding, or By recognizing it in the contract In the latter case, caution is advised since experience shows that this strategy will only be cost-effective if the contractor has proper and effective control over the source of the risk or risks concemed. Contingent planning includes: The management of a contingency budget The development of schedule altematives and work-arounds Complete emergency responses to deal with major specific areas of risk An assessment of project shut-down liabilities In complex situations, the effects of a11 such strategies can, if required, be analyzed by making appropriate changes to the risk model suggested in Chapter IVB.3. In this way decisions can be optirnized and the project can proceed with increased confidente.

C. Insurance
Those types of risk which are insurable can be selected from the list of possible risks to the project, and duly insured. For example:'
1. Direct Property Damage

.

resulting from auto collision or other auto events to equipment, in transit or handling, etc. to project materiais, including theft to contractors' property cost of removing direct loss debris equipment replacement rental income loss business intenuption iiquidated damages increased financing public bodily harm property damage arising from the negiigence of others personal injury arising from the negiigence of others damage to the project entity due to design errors execution errors project failure to perform as specified

2. Indirect Consequential Loss

. . .

3. Legal Liability

E. Data Collection, Application and Documentation
1. Historicai Databases

4. Personnel-Related

employee bodily injury cost to replace employee resulting business loss

Project risk management, particularly the risk evaluation and analysis activities, is data intensive. A reliable data source or sources is essential. Very often the required data is simply not available, and other techniques have to be adopted to simulate the particular risk or risk groupings under consideration. However,

Chapter VI

Risk Response and Documentation
even when data is available, a practical difficulty is the correct interpretation of the explanatory descriptions when applying it to anticipated risks on current work. Just as with project cost estirnating, there is no better source of information than an organization's own historical database. This should consist of recorded risk events and experience on past projects, preferably of a similar nature where these exist, built up as a result of formalized post-project assessments described below.
2. Current Project Database

Risk Management

Chapter VI1

Management of Contingency Allowances

If a project is already in the implementation phase, every effort should be made to coliect appropriate ongoini data to establish a current projecf database of frequently recurring risk, as the project proceeds. This will be particularly valuable for updating the assessment of overall project risk. However, for many of the risks, especiaiiy at the initial impact analysis stage, the data are necessarily subjective in nature and must be obtained through lessons learned by others, or by careful questioning of experts or persons with the relevant knowledge. For example, an expert could be asked to estimate the optimistic, most probable and pessimistic values for a particular variable, together with explanation of the factors which might contribute to the degree of variability represented in these estimates. The amount at stake for each and the sensitivity of this amount to changes in related variables must also be determined.
3. Post-Project Review and Archive

The compilation of useful historical data is quite a chaiienging task. It is a task which may be set aside unless its value is recognized and the data coilected and organized as part of the project's ongoing management responsibiiity. The descriptions of risk assessments, events experienced and their consequences should all be recorded. Consistently structured post-project reuiews or evaluations in which planning assumptions are matched with actual experience, vanances conscientiously explained,and the overall success of the project thoroughly assessed, are essential to consistently good project management. All essential data associated with risk management, systematicaiiy coiiected, carefully structured, and accurately recorded while each project is ongoing, should also be included in the post-project review. The key elements of each post-project review should then be abstracted and appropriately archived to update the organization's histoncal database. The post-project review and data archiving should be considered as part of the project's termination phase and completed before its conclusion.Such documentation can be quite a demanding technical responsibility. 1. PMBOK, 3/28/87, pE-4.

A. After the Known Risks, What Then? So far this handbook has dealt with the identification of project risks and how they can be evaluated. That's fine for the risks that can be identified, the known-unknowns. But what about the unknown-unknowns, should they be ignored? Maybe, it's a question of interpretation. Given the definition of unknown-unknowns in Chapter III.C, it is simply not possible to plan for unknown-unknowns. The 1973 Arab oil embargo is a good example of an unknown-unknown. Cost estimates which were prepared prior to the embargo contained no provisions for such increased fuel costs-no estimator could have possibly been expected to foresee that problem. In other words, in looking into the future, there is absolutely no way that unknown-unknowns can be taken into account. We must simply take our lumps as they materialize.' However, problems do surface on projects which are "unexpected" and are tagged with the inference of being unknownunknowns simply because they were not thought of during planning (even though they should have been!). In reality, they are known-unknowns whose existence and degree of impact on the project are uncertain. Experiencing "unexpected" events may be a reflection of weaknesses in the original planning assumptions. Planning assumption weaknesses may include: Predictions that prove to be false or overstated, Lack of timely information for project perfomance, Loss of control during project realization, Human resources that are not available when required, Lack of competence in key individuais or workforce, Disruption due to personal conflicts or interna1 politics, or simply, The occurrence of misunderstandings, burn-out and sickness.
Any such weaknesses should be carefully examined and accounted for by suitable mitigating steps, contingency plans, work-arounds, or otherwise accommodated from allowances for identified risks. If, however, the impact is greater than can be so accommodated, then the project manager may well have to negotiate a change in the scope or

Chapter VI1

Management of Contingency Allowances qualityof the work, take more time, spend more money, or some of a11 four, in order to complete the project. If a situation develops which is beyond recall, then the project should be aborted as promptly as possible to minimize further wasted time and effort.

Risk Management

B. Contingency Reserves? From the foregoing discussion it would appear to be pmdent to hold some contingency reserves, where this is permissible, to cover unexpected needs on the part of the project sponsor, principally in scope and quality requirements, which will surface in both the time and cost dimensions. However, the existence of any such allowances and their proper management present the project sponsor, and indeed the project manager, with some difficult philosophical and psychological management choices. For example: should they be added to the estimates of time and cost of the project? Or should some flexibility be permitted in the accomplishment of the scope and quality of the project instead? Neither the client nor the project manager relishes having to make changes to the project's constraints once they have been agreed upon. Wise project managers negotiate time and cost allowances in their plans from the beginning to provide them with some management flexibility. Sometimes this flexibility can be effectively accomplished by identifying some peripheral or minor scope objectives that are not mandatory. Such flexibility greatly enhances the project manager's ability to exercise influence in managing the project. If time and cost reserves are set aside, should they be under the control of the project's sponsor, the project manager, or the functional managers? Should they be global like big pots to be dipped into when needed, or should they be allocated on some basis appropriate to authority and responsibility on the project? And finally, having regard to the propensity for any spare time and money to be spent, often well before the project is even completed, should such allowances be public or secret? A prudent sponsor will hold some schedule and financial reserves for the sponsor's own needs, and the project manager rnay or rnay not be aware of their existence. An "unsophisticated" client, on the other hand, rnay not know enough to have them, so the project manager should not count on access to any "secret" reserves. In practice, the answers to these issues probably lie in the strength and experience of the project's sponsor, management, and general culture surrounding the project. What these questions do point to, however, is the importance of applying risk management, first to reduce project risks to an acceptable minimum, and then to justify effective cost and schedule contingency allowances to cover the remainder. It will be u p to project management to properly manage and control these allowances for the ultimate success of the project.

C. Application of Project Contingency Allowances Contingency allowances are different, separate, and in addition to the schedule and financial resources determined by good estimating techniques. Good estimating requires stating the estimating strategy, the planning assumptions and the typical nsks included. For exarnple, when using a probabilistic estimating technique, partidariy when a range of any type is associated with the cost, the given range is adually a "basiccontingency"allowance,both positive s and negative, for that cost. Basic contingency i designed to cover the inherent variability in the cost of the given element. For that reason, in probabilistic estimates, no additional "basic contingency" should be added at the bottom of the estimate. That would be accounting for it twice2 Separate from the "basic contingency" issue, there rnay be a need to consider other types of allowance to cover recognized scope uncertainties, resource uncertainties, and so on. An actual contingency amount is then derived from these adjustments which are needed to bring the project's probability of ovemn or underrun to an acceptable level. Such allowances should be applied just once-at the "bottom line" of the estimate. They should not be duplicated or "layered as a result of various estimate segmentsbeing combined to form larger segments.Othenuise, the accumulated total rnay be excessive and make the overall project plan too long, too costly, or both3 D. Contingency Allowances for Project Implementation Strategies for handling contingencies in the implementation phase of a project, that is, after the golno-go decision, will likely depend on the type of project, its criticality, and whether it is being conducted with resources interna1or external to the organization. In any case, for effective schedule and cost control during project implementation, a reaiistic project schedule and budget must be approved as the baseline terms of reference. This approval normally depends on the schedule and cost estimates submitted as part of the output of the project's development and planning phase. For purposes of the following illustrations,we will lookat the stages of a construction industry project. Approaches to other types of projects should be adjusted accordingly. It will be assumed that the project will be realized substantially by external resources under contract. If this is not the case, then the approach should be modified to suit the particular in-house policies and procedures of the sponsonng organization in question. For our purposes in amving at appropriate schedule and cost contingency allowances for the implementation phase, this phase rnay be viewed as having three stages as follows4
i Before contract award. In this stage, after approval to proceed .

with the project but before award of any work to contractors, it will be necessary to estabiish the scope and quality of work

Chapter VI1

Managernent of Contingency Allowances
required in as much detail as possible. This is typically conveyed by means of scope of work descriptions, designs and specifications.This work itself may be the subject of contracts, if not performed intemally. Either way, the quality of this work is critica1 to subsequent success and requires sound project management, with regular reviews until the product of this stage is satisfactory. 2. Contract award. This is the procurement, tendering and award stage. When placing an order or contract, especially if firm price bids are being sought, a variance wiil inevitably exist between the estirnates developed pretender and the schedules and prices submitted. This is due to market pressures at the e time of bidding, such as s e ~ cand material price fluctuations, impending labor agreements or disputes, productivity assessments, and whether contractors are busy or slack and their consequent attitude to risk. 3. After contract award. It is in this stage, the main production stage, that the various unforeseen items reaily emerge. Typically, this is evidenced by the necessity for issuing change instructions. Any change will have a disruptive effect on the work. The more changes, and the later they are, the more disruptive to schedule and cost will be their effect. In many ways, the number of changes that are necessary after contract award will be a reflection of how well the planning was done prior to award, or the detaiiing of the work earlier in the current phase. Of course, the sponsoring organization may have set a deliberate policy of awarding contracts before aii the information is complete in order to reduce the overaii time for the project. This project management strategy is often referred to as "fast-tracking," and puts the project into a much higher risk cate gory. Under these circumstances, appropriate contingency allowances must be increased accordingly.

Risk Managernent
risk event. The concept is based on probability "swings-androundabouts" and so it is most important to establish policies and procedures for managing the complete contingency allowance. A recomrnended approach is that once appropriate allowances have been determined, ailocations should be made to the major functional areas of responsibility associated with each of the three implementation stages described i Section D. Table VII.1 n shows a matrix chart for a construction project in which allocations have been made to each of these three stages and to each of four functional areas of responsibility.
Table VII.1. Contingency Allocation: Construction Project

STAGE RESPONSIBILITY PRE-AWARD Policy Changes Scope Changes Enhancements (All nominal) AWARD POST-AWARD Policy Changes Scope Changes Enhancements 112% Budget and Cash Row Dicuities Tendering Variances 5% Market Conditions Inflation Other Contractors 1 12 % Equipment Delivery Coordination 'Making it Work'
12%

OWNER

1%

MANAGER

Schedule Delays

%

DESIGNER

Re-evaluationof Design 1% Design Changes

CONTRACTOR

Changed Conditions Claims for Delays % Inflation Legal Interpretation

,

E. Implementation Contingency Strategies Typical approaches to setting contingencies vary from applying standard allowances, to percentages based on past experiente, to a careful assessment based on the sum total of the most Iikely probability and consequences of the various risk items properly identified.' It is the latter which is being advocated in this handbook, as part of a pro-active and responsible approach to project risk management.A simple forrnat is described in Section F. In fact, the first two approaches may be characterized as the "big pot" approach. When anything goes wrong, everyone dips into the big pot. Unfortunately,the pot always seems to be emptied long before the project is completed. Indeed, the reality is exacerbated by the perception, because people are tempted to raid the pot before it is too late! It is true that when a contingency aiiowance is based on probability and severity of occurrence and is allocated to each item proportionately, no singleamountis sufficientto cover its corresponding

In the example, suppose a total contingency cost allowance of 10 percent for the irnplementation phase has been arrived at. Then, say, 1 percent might be aiIocated to each of the sponsor, project manager and designers in the preaward stage for a total of 3 percent, and 5 percent allowance assigned to the award stage. This leaves 2 percent for the post-award stage, plus whatever is camed forward from the two earlier stages. T i latter carry hs forward is justified as follows: Assuming that the original estirnate is reasonably sound, and that firm price tenders have been called for, then if the market conditions are depressed, very competitive bids may be received. This could mean that little, if any, of the award stage allowance is consumed; indeed, it might even be enhanced. However, it is then appropriate to increase the post-award stage contingency allowances because the probability of contract administration difficulties and consequent claims arising is correspondingly increased.

Chapter VI1

Management of Contingency Allowances

Risk Management

Of the three stages described, contract award is probably the most uncertain. For example, in construction work, even under good bidding conditions and on a "good set of documents," the core bids (i.e., those bids not containing gross estimating or computation errors or extremes) are Likely to vary at least by 5 percent to 10 percent, simply as a result of differences in production methods, productivity assumptions and required margins. However, if the bids are scattered well beyond this range, perhaps the documents should be re-examined for ambiguity or lack of clanty and precision.

Chapter VIII

Managing the Risks of the Project's Environment

F. A Simple Tabular Calculation A simple computation to amve at an appropriate estimating contingency allowance to cover the consequences of second- and lower-order nsks may be amved at by using the tabular format shown in Table VII.2
Table Vii.2. Simple Tabular Calculation of Estimating Contingency

Description of Risk Event Risk Event #1 Risk Event #2 etc.

Probability of Occurrence Probability P do. do.

Estimated Cost Risk Event Status of Consequence (Criterion Value) (Amount at Stake) $ Cost C do. do. PxC do. do.
C PxC

Project Estimating Contingency based on:

A. What is the Project Environment? Why wony about the project environment...when the objective of project management is to get the project completed within scope, cost and schedule? The truth is that what is ultirnately at risk is project success, so important though these objective criteria are, they are not necessariiy the ultimate determinants of success. Heresy? Perhaps. But success, a very elusive notion at best, is dependent upon navigating the project through all its various uncertainties (risks)and ending u p with satisfied customers. Who are the customers?In varying degrees, just about anyone n involved with the project, i other words, the project stakeholders. It is these stakeholders collectively who establish the various cultures surrounding the project and thereby establish the project's environment. Two views of the propd environrnent are shown conceptually in Figures VIII.1 and 2. The foiíowing discussion

This exercise itself may well identify higher-than-reasonable risks requiring risk management focus. All first-order risks, i.e., those with both a high probabiiity and high amount-at-stake should, of course, be reported to senior management or the project sponsor for appropriate action. Depending on that decision, and the degree of controiiabiiity of the particular nsk, it may be necessary for project rnanagement to establish a disclaimer regarding the consequences. In summary, effective contingency management requires a positive and systematic approach. By doing so, changes in the status of the dowances and, hence, the trends in the final position at the end of the project will be very good indicators of its health. Moreover, adverse trends can be brought to the attention of those who are able to significantly influence the future course of events in the interests of the project's ultimate success.
1. M.W. Curran, Decision Sciences Corporation, in letter dated November 26,1990. 2. Ibid. 3. A.M. Ruskin and W.E. Estes, What Evenj Engzneer Should Know about Projed Managemmt, Marcel Dekker, Inc., New York, 1982, p45. 4. RM. Wideman, Cost Contml of Capital ~r&xts, A.E.W. SeMces, Vancouver, B C , . . 1983, p75. 5. ibid., p76.

a o o-+ o
Regulators

Suppliers of Inputs

Consurners of Outputs

Cornpetltors

Figure Vm.1. Projed Environment Process
Aiter N.R. Bumeti and R. Youker. EDI Training Coune. CN-848. July 1980.

VIII 1

-

Chapter VI I1

Managing the Risks of the Project's Environment

Risk Management
This is particularly tme of hard, comrnercial and industrial, projects, but it is alço often tme of soft, administrative-type projects, and the subsequent stress in the working environment. The problem is that people are understandably resistant to change because the known is always more comfortable than the unknown. For example, when a situation has remained stable for a significant period, individuals develop a11 sorts of habits and short-cuts that work to their personal advantage. Change threatens to destroy this personal comfort base, forcing people to start afresh. These vested interests, together with an initial lack of understanding and a perception that there will be winners and losers, all too often constitute serious hurdles to be overcome. Yet too frequently these very hurdles are simply overlooked during the course of the project. This resistance to change rnay be evident amongst many of the stakeholders. Others rnay simply have counter vested interests, or personal or group agendas, which are only indirectly rela ted to the project. Together with the project's sponsors, owners and users, these people constitute the project's direct and indirect stakeholders. Collectively or separately, their attitudes and expectations rnay well undermine the success of the project. This leads to the possibility of influencing these attitudes and expectations in such a way that the change which the project is designed to introduce will be much better received. If these uncertainties can be uncovered in good time, they can be dealt with pro-actively so that the corresponding risks to the project are significantly reduced.

Organization

Human Health

Figure ViIi.2. Project Environment: Externa1 Mluences and Interfaces
R.M. Wideman. Project Management Course. 1990

looks at the environment from the perspective of risk. For convenience, this environment rnay be thought of in terms of the interna1 project culture, the sponsors' (permanent) corporate culture, and the extemal social surroundings. The interna1 and corporate stakehoiders are the easiest to recognize, and the manager of the successful project understands the need to be attuned to current organizational, social and political maneuverings to avoid some classic project risks. To advantage, he or she rnay spend some time and effort i influencing n these environments for the benefit of project team performance. For example, every project team member needs to be convinced of the project's worth and become dedicated towards its objectives. Top management needs to be convinced of the benefits of appropriate training, and its power in developing competence and commitment. In this way, misunderstandings, mistakes, and rework rnay be avoided, and the consequences of such risks correspondingly reduced.

C. Risk and safetyl
A good example of the potential for attitudes to undermine a project is public concern over safety and/or environmental damage. An organization rnay have a finely tuned project risk management program in place, and still suddenly find itself in the middle of an hysterical local community distraught over public fears and media distortion of a risk issue. The fact is, risk management is not complete without effective risk communications. Suggested mechanisms for public communication are outlined in Sections E and F below, but perhaps it is worthwhile discussing at this point why traditional responses with factual data tend to fail. Research into the psychology of risk perceptions by U.S. psychologists Paul Slovic and Vince Covello indicates:
i.

B. Problems are Caused by People The extemal stakeholders are more difficult to idenhfy and include aIl those who w i l in any way be involved with the resulting entity or produd upon its completion.Typically that's a lot of people, and today there is a rapidly growing awareness and concern arnongst thern for the impads on their physical enWonment.

People do not, in fact, demand zero n s k They take risks every day, both consciously and subconsciously, and they are willing and able to take benefit/riskdecisions, as in driving and speeding. Some take deliberate risks simply for pleasure, prestige or self-satisfaction (e.g., sky diving). 2. Peoples' judgments of degrees of risk are not, however, coincident with most methodologies for measuring risk statistically. The public rnay greatly underestimate familiar risks

Chapter VIII

Managing the Risks of the Project's Environment

Risk Management

(e.g., driving) while greatly overestimating unfarniliar risks (e.g., buying a home near a nuclear facility). 3. A variety of emotional, not logical, factors control risk perceptions:

public emotional response to risk, and responding by seeking a change in the attitude towards those who are being held responsible for creating and managing the risk. This is i contrast to n seeking a change in the perception of the risk itself.

. .

Primary is the sense of personal control, i.e., the ability to choose taking a risk (iike getting into a car) and the ability to manage the risk ( I am the one behind the wheel). Secondary are qualities of familiarity and, conversely, dread. The greater the unfamiliarity and potential for connection to gruesome (e.g., nuclear impacts), the more it is likely to be judged as highly risky and therefore unacceptable.

4. Once established, risk perceptions are extremely hard to

change. New information may be absorbed by the intellect but it is not readily absorbed at an emotional level. 5. Risk perceptions reside fundamentally at an emotional level. What these insights suggest is that, in a crisis of fear, the traditional management instinct of providing rational, statisticaily-based information could be misguided. In a recent egg/salmonella case, spokesmen for the egg producers tried to explain the extremely low risks of an individual actually being made sick by an infected egg. What they failed to recognize was the sanctity of food and the trust in those who deliver it. It is not a question of how much risk, but rather the sudden imposition without warning of risk into a system assumed to be low- risk. As a result, the problem effectively got out of control. In the exarnple, risk communications should have tried to address the issues of surprise ments and loss of control. Instead of trymg to prove that there was "no problem, except for the elderly, the infirrn and babies," a better response might have been "A detailed exarnination has been ordered to determine the extent to which the problem exists. We will get to the bottom of it and keep you fully informed. For the moment we recommend that a11 eggs be thoroughly cooked, and that the elderly, infirrn and smaii children limit their consumption. Anyone with these symptorns should imrnediately call, etc., etc." The differencein the two approaches is that the first assumes that there is a basis of trust, or faiis to recognize the need to establish a basis for trust by asking people to be reasonable. The second endeavors to establish a basis of trust by showing that the problem is being actively monitored and controlled. If the research demonstrates that the fttndamentais of risk perception are emotional and not rational, then the primary focus of project risk management communications should be to estabiish (or re-establish) trust in the organization, rather than to educate the public about science or technology and its benefits. This means beginning by recogruzing the natural legitimacy of

D. Principal Determinants To identify the potential difficulties associated with project risks and to assess their probability of occurrence, designated members of the project team must interact frequently with those institutions and individuals which constitute the most important elements of the project's stakeholders.The required effort can then be priorized with a view to heading off the most serious obstacles well in advance. This environment will not be the same for every project, of course. In fact, it is iikely to be detennined principally by three considerations:2
The product or semice resulting from the project, The technology and the manner of its application, and Its physical location. Suggested steps in the process of identification are:3
i.

2.
3. 4.

5.

6.
7.

8.

Leam how to understand the role of the various stakeholders, and how this information may be used as an opportunity to improve both the perception and reception of the project. Ident* the real nature of each stakeholder group's business and their consequent interest in the project. Understand their motivation and behavior. Assess how they may react to various approaches. Pinpoint the characteristics of the stakeholders' environment and develop appropriate responses to facilitatea good relationship. Learn project management's role in responding to the stakeholders' drive behind the project. Determine the key areas which will have the most irnpact on the successfulreception of the project by all the stakeholders. Develop a Project Acceptance Plan aimed at rnanaging externa1 stakeholders' interests.

However, remember always that even a minor stakeholder group may discover that "fatal flaw" in the project's concept that could bring it to a standstill! Failure to deal with these issues in a timely manner will inevitably lead to a less than optimum project outcome.

E. hnanaging by Stakeholder Groupings and Categories Project stakeholders may be recognized in any of the following categories:4 Those who are directly related to the project by having a stake in its process or product such as suppliers of inputs, managers of the process, or consumers of outputs; VIII 5

VIII 4

-

-

Chapter VIII

Managing the Risks of the Project's Environment Those who have influence over the physical, infrastructural, technological, comrnercial/financial/socioeconomic, politor ical/legal conditions; Those who have a hierarchical relationship to the project such as government authoritiesat.local,regional and national levels; and Those individuals, groups or associations, sometimes only indirectly related to the project, but who see a linkage between their own interests and the project's goals and who pursue it for their own ends. These may include special interest groups or competitors, but also those providing services who wish to profit at the expense of the project.
Having identified the various stakeholder groupings, each may be assigned to a category according to their amenability to influence. Three categories are suggested, namely: Those who are controllable, Those who are influenceable, and Those who need to be appreciated. Within each category, stakeholders may then be further rated by degree of importante according to their ability to influence the outcome of the project. Members of the project team can then pnoritize their efforts to maintain effective stakeholder linkages, designed to give the best chances of ultimate project success. Possibly the largest constituency are those who are relatively neutra1 at the outset and can therefore be influenced in their opinion. They may also represent the best source of potential support and hence the best opportunity to establish a perception of project success, if their opinion is mobilized effectively. If the project is sufficiently large or visible, a separate program of effort may be assigned to a specific group as a special public relafions(PR) eff ort. Either way, the trick is to persuade the various players that what they want is the same as the project's objectives, or else to modify the project's objectivesaccordingly. The goal is to establish such a congruence through risk management.

Risk Management
Genuinely sincere appreciation expressed to inquirers Flexible personal responses provided, where special issues dictate Recovery from inevitable lapses of existing services during project activities, in ways that impress Project team members empowered to make decisions to solve urgent and obvious local problems Stakeholder-friendly information faciiities available both during project implementation, as well as subsequently Traditional management has long since recognized the value of the classic input-process-output model, with its management information feedback loop for controlling output. Dynamic managers have also recognized that opening communication channels in both directions, such that modified management strategies result, can be a particularly powerful workforce motivator. The key is quaiity information. Whether this information is presented in verbal, written or graphical form, improvements in performance can be quite remarkable and, indeed, many "knowledge workers" demand it to provide them with job satisfaction. The concept is shown in Figure VIII.3. The principle is just as true in the field of projects, though regretfully less evident, especiaily where the externa1 environment is concerned. Thus, the project manager's job is no longer confined to controlling events within his or her own organization. It is no longer sufficient to think of project management simply as the monitoring of time and cost by planning, scheduling and resource leveling, as some scheduling software promotion seems to suggest. Nor is it sufficient JLIS~ to include the many other administrative tasks required of the project manager as leader of the project team.

Positive Feed Forward

F. The Means and Value of Exercising Positive Influente What are the h a h a r k s of a successful PR program? Here is a topten check list:'
Quality information about the benefits of the project Participation in those decisions which have some degree of flexibility, or constitute viable alternatives, and which may result in enhanced project value Care and concem genuinely expressed for the project's stakeholders Information requirements anticipated and provided ahead of time Timely response to other information requests

Public

c
Public Feedback
Figure V i . . Managing the Environment: Public Relations Concept II3
R.M. Wtdeman. Managing the Project Environment. Dimensions of Project Management, Springer-Veriag. New York. 1990. p63.

VIII 7

-

Chapter VIII

Managing the Risks of the Project's Environment
Today, to a remarkable extent, the success of a project depends on the ability of the project manager and his or her team to establish a genuine and positive infiuence over the attitudes of the stakeholders. A favorable perception means a successful reception!

Risk Management

Chapter IX

Dealing with Risks in Contracts

A. Contract Strategy Considerations Selection of an appropriate procurement strategy will depend upon the type of project, its particular emphasis in terms of scope, quality, time and cost, and the degree of uncertainty associated with each. Careful consideration of these aspects should lead to the right choice of organizational structure, allocation of responsibility, and means of procurement. If the project is to be realized by intemal resources, then ãny commitments will be less formal since they are not "at arms length in a strictly legal sense even though they may become part of an employment contract. However, any risk. arising out of uncertainty under agreements internal to the organization rest with the organization, so the issue is not "contractual" but rather one of intemal policies and procedures. Whenever any work is to be contracted for outside the organization then a suitable contract needs to be formulated in some detail. The manner of its development, tendering or negotiation and award procedures, are the subject of the PMI Handbook on Contract/Procurement. However, a range of types of contract is possible in which thedegree of associated risk is shared dilferently between the parties. The choice of which type of contract is most suited to the project in question is very much the subject of project risk management. The selection of the right form of contract requires:
The identification of specific risks, Determination of how they should be shared between the parties, and The insertion of clear legal language in the contract documents to put it into effect. 1. Adapted from J. Lindheim, Distress Signals, Management Today, BIM, London, April1989, p105. 2. Galbraith, 1977. 3. R.M. Wideman, Good Public Relations: An Essential Part of Successful Project Management, PMI Seminar/Symposium Proceedings, Denver, Colorado, 1985. 4. Adapted from Asian Development Bank, Analyzing the Project Environment, 13thADB Regional Seminar MS, Manila, Philippines, 1987. 5. D.M. Connor, abstracted from promotional literature, Connor Development Services, Victoria, B.C., 1989. The most challenging of these tasks is the finding of a cost-effective and equitable degree of risk allocation. Standard contract (or purchase order) documents prepared by various leveis of governrnent, organizationswhich undertake a lot of procurement, or standard model documents prepared for various industry sectors, such as construction, are typically used. Specific allocations of risk are intrinsic to such standard forms, but the principles behind the allocations are rarely stated.' Such intrinsic ailocation of risk may or may not be appropriate to the project.

VIII 8

-

IX- 1

Chapter IX

Dealing with Risks in Contracts

Risk Management

B. Suggested Risk Sharing Principies Various authors have sought to identify principies which should overn the allocation of risk arnongst the parti- to a project.8 Recommendations are suggested by the answers to these questions:
Which party is the source of the particular risk and hence best able to control the events that may lead to it happening in the first place? Which party can best manage the risk, if it occurs? 1sit or is it not preferable for the client to retain an involvement in the management of the risk in question? If it cannot be controlled, which party can or should cany the risk? Wiil the cost incurred, or premium charged, by the recipient of the risk be reasonable and cost-effective? Wiil the recipient be capable of sustaining the consequences of the risk, if it occurs? Or will ! lead to the possibility of other risks of a different i nature being transferred back again, e.g., slow down or collapse of the project effort through demoralization or bankruptcy, etc.? If at the time of entering into a contract the distribution of risk is not clearly understood, or is patently unfair, such as one party being the cause of risks which are then sustained by the other party, thendisputes arealmost inevitable. Theadoption of systematic project risk identification and analysis may consequently result in departing from the standard contract conditions in particular circumstances, to the overaii benefit of the project.

C. Types of Contract
The t e m contract has been defined in various ways by various authorities. Alço, the law of contract depends on the jurisdiction, so that the essential attributes of a valid contract may vary. However, for project purposes, two requirements appear to be irnmutable-in return for a legal act, there shall be legal "consideration." It is the consideration, or form of payment mechanism, which has led to a variety of different contracts. These different types of contract range from fixed-pricelump-sum for the identified work, to fully cost-reimbursable. While the payment mechanism may be a reflection of the current practice in a particular industry, it is also often a reflection of how much is known beforehand of the work to be accomplished. Work that is fully described, can proceed without intermption and can therefore be scheduled and estimated, is appropriate to a fixed price. Work that is essentially undefined and/or cannot be scheduled is obviously highly uncertain, and therefore risky, and in fact is unlikely to proceed except on some form of cost reirnbursable basis. Figure IX.1 shows a typical relationship between different types of contract and the allocation of risk. In other words, the payment mechanism is a reflection of the degree of uncertainty, i.e., risk. So it transpires that one of the most significantdifferencesbetween the various forms of contract is the way in which risks are handled and p i d for, ranging from those which are pie-based to those which are cost-based. 1n price-based contracts (i.e., lump sum, and, to a slightly lesser extent, unit price contracts where the quantities of individual units are expected to vary), payment is deemed to cover all costs, overheads and profit, wherein the contractor includes all necessary contingencies for risks specified and/or inferred. Cost-based contracts are at the other end of the scale. Here, all costs actually innirred in the performance of the contract are reimbursed and, in addition, a fee is paid to cover profit and any associated costs and overheads which are not defined as reimbursable. It follows from this arrangement that virtually a11 risks are paid for by the client, unless specified otherwise and covered by the agreed upon fee. Almost as a corollary to the aiiocation of risk between these two extremes is the degree to which the owner or client has the opportunity to get involved in the management and organization of the work and to make changes. In the former, management and organization of the work must be in the hands of the contractor with quite rigid procedures for making changes. In the latter, there can be a high degree of flexibility. At first glance, it might be supposed that the cost-reimbursable approach is the ideal form of contract. Unfortunately, it is the subject of very detailed and accurate cost and progress record keeping, which even then may be the subject of independent audit and dispute. Perhaps more important is the lack of incentive to control cost and schedule by those engaged under the contract.

SUGGESTED

ALLOCATION

CPPF Cost Plus Percentage Fee CPIF Cost Plus Incenlive Fee

CPFF - Cost Plus Fixed Fee

FPPI Fixed Price Plus Incentive FFP - Flrm Fixed Price

-

Figure DC.1. Contract m vs. Risk Allocation e

Chapter IX

Dealing with Risks in Contracts

Risk Management
Table IX.l. Risk Implications of Different Types of Contract (from Client's perspective)

This has given rise to a whole range of cosf-plus-incentive types of contract. The range of possibilities tend to be a function of the imagination of the respective contract negotiators and seem to be alrnost endless. Inany case, this is the subject of Contract/Procurement Management. Suffice it to repeat here that the particular form of contract adopted will be a reflection of the degree of uncertainty, and how the nsks are to be aíiocated and paid for.

Issue Financia1 objectives of client and contractor

I

Type of Contract Lump Sum Different but reasonably independent Unit Price Different and in potential conflict Target Cost Cost Reimbursable Both based on actual cost but in potential conf lict

D. Different Contract Risk Implications Figure IX.2 shows probability curve distributions for three projects ali with the same expected value. However, the probability of project A being completed for this value is the highest compareci to projects B and C because it has the best definition of scope of work. The figure alço suggests the type of contract appropriate to the range of uncertainty. Table IX.1 shows the impact of different issues on the different . also shows how nsk may be distributed or types of ~ o n t r a c tIt~ handled in each case. E. More Than One Contract In many project situations, more than a single implementation contract is involved. In program management, for example, there will be a senes of projects, running either consecutively or concurrently which will no doubt involve separate contracts. Or within a project, particularly one which is complex, large, and/or

Considerable harmony. Reduction of actual cost is a common objective provided cost remains within the incentive region

Contractor's involvement in design

Excluded if competitive price based on full design and specifications Excluded

Usually excluded Contractor encouraged to contribute ideas for reducing cost Virtually excluded Possible through joint planning

Contractor may be appointed for design input prior to execution

Client involvement in management of execution Claims resolution

Should be active involvement

Very difficult, no basis for $$ evaluation

Difficult, only limited basis for $$ evaluation

HIGH

Project A Well defined scope and work content. High probability of achieving realistic cost estimate at 100°/~

Potentially easy, based on actual costs. Contract needs careful drafting

Unnecessary except for fee adjustment. Usually relatively easy

II \
PROBABILITY OF OCCURRENCE MEDIUM

f Proiect

B

Fairly well defined scope and work content. Fair probability of achieving

I

Forecast final cost at time of bid

Known, except for unknown claims and changes

Uncertain, depending on quantity variations and unknown clairns and changes

LOW

k
80% 90% 95% 100% 110% 120% 140%

Poorly defined scope and content. Low probabiliiy of 100% cost estimate Payment for cost of risk events Depending on contract terms, undisclosed contingency, if any, in contractor's bid. Otherwise by claim and negotiation Depending on contract terms, undisclosed contingency, if any, in contractor's bid. Otherwise by claim and negotiation

Uncertain. Target Unknown cost usually increased by changes, but effective joint management and eff icient working can reduce final cost below an original realistic budget Payment of actual cost of dealing with risks as they occur, and target adjusted accordingly Payment of actual costs

COST ESTIMATE VALUE Suggested types of contract for various spreads

I
-

I L +I- 15%:- FFP J
+ - :- CPFF I 25% +/- 50% .- CPIF

I

I

> 50% :- CPPF

Figure IX2. Scope Definition Risk - Contract Selection

Chapter IX

Dealing with Risks in Contracts
organized on a specialty or trade basis, there rnay be numerous separate service agreements and contracts to be coordinated. The work involved in the different phases and stages of the project rnay also be the subject of different contracts or agreements. For example, in the project development phase, social, environmental, market and technical studies rnay be procured from different specialists. Even a separate risk analysis study rnay be commissiond for a large or critical project. In subsequent stages detaiíed specification, design and supervision or quality control and expditing rnay be the subject of separate specialized packages, and so on. The greater the number of separate packages under separate and discrete responsibilities, the greater the amount of coordination required and the greater the resulting risk Consequently, early in the development of the project the project sponsor must develop a suitable contract strategy and establish an organizational structure consistent with the intended project objectives. The organizational structure selected must reflect the proper division of responsibiiity for these various components in terms of scope, quality, time and cost, including any delegation of project management responsibility. The strategy must recognize the need for management coordination and integration of the various contracts over time by making corresponding provisions in each of the contracts. There are many different ways of stmctnring a project's procurement to suit differing objectives and circumstances. Different structures assign risk in different ways. Perhaps the most significant issue for project risk management is the overall distribution of responsibility for project risk as reflected by: The ways in which management responsibiíity is structured, How it is delegated, and How it is incorporated into the various contracts. Responsibility for the costs of risk events for different types of typical standard contract is also compared in Table IX.1.

Risk Management
Clear definitions of risk and their assignment Positive incentives linked to risk assignment Flexibility for different assignment of different risks between parties Strong emphasis on good management practices designed to avoid unnecessary risks Handling project risk in the way described rnay require a departure from sometimestraditional forms of contract. However, if the effect is to avoid or reduce the number of contract disputes, as well as the amounts of money under dispute, then the respective levels of management will have more time to spend on the real issue-project success.

F. A Question of Attitudes Many descriptions of successful projects (e.g. Showcase Projects reported in the PM NETwork) reflect on the healthy cooperative t e m spirit enjoyed on the project. So it is not necessarily the project organizational structure and f orms of contract which determine whether or not project objectives are successful1y achieved, but rather the attitudes of the parties involved. However, an effective structure and good contract wording can go a long way to establish good relations and avoid the frustrations which othenvise undennine initial enthusiasm and good intentions. Obviously, this includes an equitable distribution of risk, and the means for handling it in the event that it arises. Suggested characteristics of good contract conditions in respect of risk in~lude:~

1. R.W. Hayes et al, Risk Management in Engineering Construction, Special SERC Report by the Project Management Group, UMIST, Thomas Telford Ltd., London, December 1986. 2. Ibid., p26. Other authors include D.B. Ashley, C.E. Porter, N.M.L. Barnes, J.G. Peny and P.A. Thompson, p30. See also Contract Risk Allocation and Cost Effectiveness, C11 Publication 5-3, Austin, Texas, November 1988. 3. After R.W. Hayes et al, Risk Management in Engineering Construction, Special SERC Report by the Project Management Group, UMIST, Thomas Telford Ltd., London, December 1986, p29. 4. Ibid., p28.

Risk Management

Chapter X

SummarylConclusions

A. Risk Management The Present Projects are launched to take advantage of opportunities, but opportunities are associated with uncertainties which have risks attached. For the project to be viable, the expected value resulting from a favorable probability of gain must be higher than the consequences and probability of loss. Therefore, the riskç associated with a project must receive careful examination in the context of the organizationlswillingness or aversion to taking risks. This is the domain of Project Risk Management, which fonns a vital part of Project Management. With careful planning and good management some inherent risks in the project management process can be substantially reduced or virtually eliminated. Steps towards this goal include:
Thorough and realistic appraisal of the project concept, and hence scope definition, in the concept phase Observing good project management practices in the project planning and development phase, including realistic estimating of time and cost for defined scope and quality Examhing theuncertainty and riskinherent in the project, and identifying ways of mitigation, and/or making due contingency allowances where the scope of work is uncertain Preparing contingent action plans and work-arounds Developing sound procurement strategies designed to optimize performance, supported by an appropriate organizational structure and responsibility distribution Assigning specific responsibility for risk in a way that motivates by recognizing that risk and reward go together Exarnining contract documents for risk identification, general clarity and potential sources of misunderstanding Seeking innovative but practical solutions to offset potential areas of risk While many of the detaiis described in the foregoing chapters of this handbook are applicable to large complex projects, the principles involved are just as applicable to any size of project in any field of endeavor. Since there is no point in taking any risk that has negative expected value, the principles of risk management should be an established part of early project management activities on ai1 projects-whether complex or simple, large or smail.

-

Chapter X Summary/Conclusions
B. It's the Attitude That Counts Successful projects such as Showcase Projects reported in the PM NETwork emphasize the benefits of a healthy cooperative project team spirit and constructive attitude. Appropriate procurement strategies, project organizational structure, responsibility distribution and corresponding forms of contract should all be designed to support this positive attitude towards successfully achieving project objectives. Effective and equitable handling of project risks can go a long way towards establishing such good relations and avoiding the frustrations which otherwise undermine initial enthusiasm and good intentions.

Appendix A

Appendix A
Typical Project Risks
1. How Does the Project Manager Know When There is a Project ~ i s k ? ' Some of the most common general project risk situations encountered: The project sponsor (and the project manager) do not recognize that every project is an exercise in nsk. This project is very different from the last one. There is a feeling of uneasiness. When the project is in its earliest phase, project risk and opportunity are highest (but the amount-at-stake is lowest). The project scope, objectives and deliverables are not clearly defined or understood. A large number of alternatives are perceived as possible. Some or all technical data is lacking. The technical process (and design) are not mature. Standards for performance are unrealistic (the best there is for everything) or are absent. Costs, schedules and performance are not expressed in ranges. The f u m e tirning of activities and events are vague. Design lacks production engineering input. Prototype of a key element is missing. There is a higher than usual R&Dcomponent. Some or aii environmental permits are outstanding. Other similar projects have been delayed or canceiied. A wide variation in bids are receivd. Some key subsystems and/or materials are sole source. No appropdate contingency plans have been developed. The project team relies entirely on the contingency allowance. Someone starts "hedging their bets"!

C. Risk Management The Future?
Because risk has a natural complexity in its variety, scenario combination possibilities and variation in probability impacts unique to each project, there are some ideal opportunities for computer applications.Obvious applications include rapid access to established databases, data storage, repetitive computations ranging from simple to complex, and sensitivity or "what-if' analyses. However, effective project risk management is alço subject to practical experience and sound judgments. Artificial intelligence (AI), expert systems (ES),and, more recently, human-computer cooperative systems (HCCS) are being developed to incorporate these added dimensions. The intent is to enhance the project manager's risk management ability to arrive at better solutions than would have been amved at without such support systems, e.g., by reducing the recurrence of similar risks. The ultimate potential of AI and ES (and HCCS) is in their ability to augment the program manager's reasoning power. But to do this a better understanding of cognition, knowledge representation, information usa e, effective decision making and risk H assessment is still needed.

-

2. Specific Project Risks
1. M.D. Martin et al, Critical Issues in the Application of Artificial Intelligence (AI) to Acquisition Management, Fonun on Artificial Intelligencein Management,Dayton, Ohio, 1988. The foilowing detaiied listings provide convenient groupings of project risks generally classified accoiding to source. Thedegree of predictabiiity and abiiity to manage appropriate response varn ies but, i any case, is independent of the risk event status (probabiiity and arnount at stake).

Typical Project Risks
External Unpredictable (and uncontrollable) a. Regulatory, i.e., unanticipated government intervention in: supply of raw matenals . environmental issues design standards production standards site location product or service sales or export . pricing special requirements b. Natural Hazards, i.e., as a result of natural elements: location storm flood earthquake c. Postulated Events, i..e., as a resuit of deliberate intent: vandalism sabotage d. Indirect Effects, i.e., occurring as a result of the project: environmental social e . Completion, i-e., failure to complete the project on account of one of the following: failure of the supporting infrastructure as a result of others failure of design, execution or supply contracts due to bankruptcy or receivership, etc. failure to provide financial support to the end of the project inappropriate project concept or configuration political unrest lack of final acceptance

Appendix A
Internal, Non-Technlcal (but generally controllable) a. Management, i.e., difficulties due to: insincerity/lack of integrity incapacity inadequacies loss of control incompatibility of goals senior staff changes inappropriate or lack of organizational structure lack of appropriate policies and procedures inadequate planning umealistic scheduiing lack of coordination inadequate project management b. Schedule, i.e., delays and time overmn due to: delays due to management difficulties above regulatory approvals labor shortages labor productivity labor stoppages material shortages late delivenes unforeseen site conditions sponsor/user scope changes accident or sabotage start-up, turn-over or launch difficulties lack of access c. Cost, i.e., o v e r m s due to: any of the schedule delays listed above inappropriate procurement strategy paynegotiations management and/or workforce inexperience lack of understanding how parts fit together contractor claims under-estimating any of the externa1factors listed previously d. Cash Flow squeezing interniption insolvency e. Loss of Potential, i.e., removal of: benefit profit Technical (and generally controllable) a. Changes in Technology rendering parts of the project obsolete parts discontinued introduced by competitors, rendering the project obsolete, uncompetitive, or unacceptable complexity introduced as a result of new technology

. .

. . . . .

External Predictable (but uncontrollable) Changes in the following are predictable, but the extent and direction i uncertain. s
a. Market Risks availability of raw matenals cost of raw materials demand, including customer/user rejection economics competition end value in the market wiliingness of buyers to honor purchase agreements b. Operational (i.e., after project completion) maintenance needs fitness for purpose safety c. Environmental Impacts d. Social Impacts e. Currency Changes f. Inflation g. Taxation

Appendix B Typical Project Risks
b. Perfonnance

quality rate of production reliability c. Risks Specific to Project's Technology in creating the entity or product . in operating or marketing it d. Design . inadequate data designer/detailer inexperience . design inadequacies detail, precision and suitability of the specification likelihood of changes during the course of the project . design vs. execution methods e. Sheer size or complexity of project

Appendix i 3
Impact Analysis Methodology
This appendix describes a risk analysis methodologyl which can be applied to the general situation of managng a project under varymg degrees of uncertainty. A risk of the complexity discussed here has generally been applied only to very large programs or projects, such as utility or infrastructure construction. The methodology is described for purposes of illustration. This type of risk evaluation should typically be carried out very early in the project life cycle. In fact, it should be camed out when real infonnation is most lacking, precisely because it is at this time that a risk analysis can be of most use to the project team in gaining an understanding of the project itself. However, in oder to achieve these benefits, the risk manage ment process should be part of the central project management pianning, not merely an optionaladjunct. The benefits will be greatest when the project team perceives risk rnanagement as a means to project ends, and not an end in itself. The depth of the process can be tailored to suit the size, nature and cimmstances of the project. The methodology which follows is not confined to front-end planning. It can be used equally well for more detailed analyses during subsequent project phases. In fact, the risk framework or models developed in the process can form a valuable baseline for subsequent detailed studies in later stages of the project. Although like most other project management functions the process of risk analysis is iterative, for simplicity it is described as a set of súc tasks in the following typical sequence:
i Problem Structuring -confirming the specific objectives of the .

. .

Legal (generally controllable)

Difficulties arising from any of the following:
a. Licences b. Patent Rights c. Contractual i.e., difficulties due to: misinterpretation misunderstanding inappropriate contracting strategy /contract type failure d. Outsider Suit e. Insider Suit f. Force Majeure

.

I. After D.T. Hulett, PMP CertificationWorkshop - Risk Management, PMI Orange County Chapter, 1991, p22.

study, and developing a framework for the analysis. - identifying a comprehensive set of risk i t e m associated with project activities incorporating judgments from many sources, initial screening of risk events as to potential risk status, and developing prelirninary risk models. 3, Risk Quantification estimating the probability of risk events and the impacts on variables of concern such as performmce, delivery, cost, reliability and impact on the project environrnent. 4. Risk Modeling - modeling the combined effects of risks within and between activities, discussing results leading to base plan adjustments for the project as appropriate, and re-running.
2. Risk Identification

-

Impact Analysis Methodology
5. Overall Economic Modeling - producing overall measures of

Appendix B
Associated with each scenario will be measures such as economics, delivery and performance impacts. Summarizing the tree will produce overall distnbutions of project cost, economic retum, etc. The point of structuring the analysis of risks in this way is to enable discussion to focus on specific uncertainties and risks in some level of detail, and provide the basis for establishing the dependency links between the various project activities. It also sets the stage for the subsequent development of schedule network diagrams for the specific analysis of schedule related risks.

variability in project economics or effectiveness as a result of incorporating the impacts of all risks considered. 6. Project Risk Report - presenting the findings and recommendations. These steps are more particularly described below.
1. Problem Structuring

Perhaps the most important steps in the process, as with any study, is to obtain agreement on the objective and a study plan for its achievement. For example, the objective rnight simply be stated as T identib responses appropriate to project risks which may impact o project scope, quality, time andlor cost. The resdting study plan wodd then involve steps, firstly, to capture the base case project scope statement in terms of project configuration, with potential project alternatives. Secondly, initiatives would be designed to force discussion of project uncertainties and risks amongst members of the project team to improve the understanding of the project. Thirdly, would be the tasks involved in the structuring of information and analyses required to support the decision making for the project irnplementation plan. The information itself will consist of two types: objective and subjective. The objective information may be b&ed on projecting today's understanding and circumstances, be derived from a similar project, be the result of statistical inference, and may lead to a large number of trails. The subjective information may have to be based on a similar project but different assumptions, different pro ects, limited observations, expert opinion, or mere guesses Three types of information structuring could be contemplated. The first is the ordering of tasks within the study itself to ensure an effective progression, but also to understand the interdependencies between tasks to ensure the necessary cross-flow of information. This breakdown typically follows the traditional areas of professional or technical expertise associated with the project domain. The second structuring is the development of a well-defined, non-overlapping work breakdown of activities for the implementation of the project. Generally these will follow the first structure, but will necessarily contain a greater level of detail, and will include design philosophies, procurement strategies and timing, as well as actual production activities. The third stmcturing relates to risk modeling such as a probability/decision tree. T i device provides the flexibility for hanhs dling different levels of complexity of risk events, including situations requiring complex conditional probability treatment. For each issue considered, the underlymg framework for the analysis is a probability tree where each brandi represents some combination of potential xenarios and their respective probabilities.

Risk identification and Screenlng Following the breakdown of the project by activity, the next step is to i d e n t e risks associated with each, in order to produce a comprehensive set of uncertainties and risks. The objective is to ensure that all such risks have been considered and discussed, employing judgments from as many relevant sources as possible. In the earliest stages of the project, concem is primarily with fundamental uncertainties, those which could force a change in project scope, or othenvise could be particularly damaging to the success of the project. At this stage, the classic "brain-storming"approach, concurrent with similar project planning sessions, could be entirely appropnate. However, a more indepth exarnination of the compounding effects of seemingly insignificant risks, such as a series of minor delays, can also be very damaging. This is especially true when the project is aimed for a particular "window of opportunity." Next, ali the risks should be coarsely screened to establish those which are:
Relevant to the project Within project responsibility for project management purposes Not otherwise covered by normal project insurance Predictable (but uncertain) More tkan of minor significance This should bring the total within manageable proportions. Where severa1 risks have broad implications, they may be conveniently grouped by source as indicatedin Appendix A, and further subdivided into discrete events (i.e., single worst cases) and timescaled events. Care should be exercised to ensure that a risk item judged minor in one case is not major in another. Once the risks have been assembled and assessed in this way, consideration should be gven to joint effects and how these can best be modeled. The level of detail wilí depend on the study objectives. For example, for the analysis of safety, where complex combinations of low probability risk events and response failures result in accidents, a detailed probability tree which allows conditional probability specification may be necessary. However, for broader study aspect., risks can usuaily be treated as dependent or independent at some percentage level. This is irnportant to identify, before considering quantification.

4

Impact Analysis Methodology
3. Risk Quantification

Appendix B

The risks identified in the previous task must now be quantified in terms of degree of uncertainty in a spectrum of certainty/risk/uncertainty (i.e., probability of occurrence) and magnitude of impact (i.e., on project objectives of scope, quality, time and cost). For this purpose, it is necessary to describe the risk together with its primary impact scenario, followed by any consequential impacts. Figure B.1 shows a conceptual relationship between risk probabilities, damage scenarios and consequences. When estimating impacts, however, it is often necessary to have a set of response decision rules in order to arrive at consistent quantification.This will depend very much on the orientation of the parti& project, i.e., whether it is primarily scope, quality time or cost driven. For example, if a delay is experienced in initiatingproject realization due to financingdelays, will the response be to accept the delay in the interests of quality and cost? Or wiU steps be taken to accelerate the work by increasing manpower (with the possibility of reduced productivity)in the interests of time? Such decision rules should be a fundamental part of the basic risk management plan for the project. More esoteric considerations cover the choice of methods to describe and combine risks for purposes of computation according to standard probability theory. The assumption that all random variables can be described by pre-selected distributions, such as normal distribution, or that all distributions are independent may greatly facilitatecalculation, but the resuits are most likely unreliable. Lack of consideration of the positive dependence that often exists between variables, such as cost and time, may result i serious n underestimation. Perhaps the most contentious aspect of risk analysis is the estimation of probabiiity distribution, due to the scarcity of relevant data. Where available, it may have to be modified to suit the project in question. Where not available, reiiance must be placed on totaily subjective estimation based on expert opinions and judgrnents from personal and/or other past experientes (see Delphi Method, Appendix C). Where insurable risks are concerned, there is a large body of existing knowledge and extensive statistical data on frequency and size of incidents ranging from natural disasters to the mancontroiled perils. These data should not be overlooked. Such information is typicaily available through insurance companies and specialists in risk management.
4. Risk Combinatlon and Modellng

1

.

A PROBABILITY LEVEL IS IDENTIFIED FOR EACH RISK EVENT MAGNITUDE

OF A PARTICULAR RISK MAGNITUDE

1 1 1 )
INCREASING MAGNITUDE OF RISK EVENT 2 . IF A RISK EVENT OCCURS, IT CAN CAUSE A NUMBER OF POSSIBLE DAMAGE SCENARIOS, EACH OF WHICH HAS A PARTICULAR PROBABILITY OF OCCURRENCE. THUS, IF AN EVENT A OCCURS, THE PROBABILíTYTHAT if WILL CAUSEMAJORDAMAGE APPEARS IN THE SHADED AREA SHOWN NONE

PROBABILITY OF DAMAGE LEVEL IF THE RISK IS REALIZED

LIGHT

MODERATE CATASTROPHIC MAIOR

INCREASING DAMAGE LEVEL

I,
PROBABILíTY OF CRERION

vALuE IF A PARTICULAR DAMAGE LEVEL IS REALIZED

I I I h
MIN
MODE
MAX

INCREASING CRITERION VALUES

Once the uncertainties and risks have been quantified within each activity of concern, the joint impact of these risks must be considered. Three levels of modeling may need to be examined, namely:
a. The joint impact of a small number of risks for detailed analysis

-

3 . A RANGE OF CRITERION VALUES IS ASSIGNED FOR EACH DAMAGE LEVEL

within an activity, b. The joint effects of a11 risks within an activity, and

Figure B.1. Conceptual Relationship Between Risks, Damage Scenarios and Consequences
After A.B. Camrnaert c. 1986

Impact Analysis Methodology
c. The overall impact of nsks from a set of, or aií, activities.
Project Direct Costs Project Indirects Interest & other charges etc., etc. Expected Value o f Risks" TOTAL PROJECT COST
$ xxx $ XX $ XX

Appendix B

At each step it is important to review and, if necessary, recycle the results, preferably with the persons responsible within the project team. The purpose is to make team members aware of the uncertainties in the estimates so that inconsistencies and weaknesses in the overall project plan can either be corrected early, or more appropnate courses of action adopted. As can be seen, these areas of nsk quantification and nsk modeling can become highly sophisticated.They then become the purview of highly specialized expertise and expenence in the particular area of project application. However, in simpler terms, a cnterion value, ranking, or status for each nsk event (or set of combined events) may be established by the following relationship:

$ 2
$XX

" Note that this rnay be a more cornprehensive and convincing way of expressing an appropriate "contingency" allowance for the project.

A conceptual nsk analysis computer program, structured for conducting the type of risk analysis described above, is shown in Figure B.2.

Risk Event Status = Risk Probability x Amount at Stake
where probabiùty3 and the amount-at-stake (i.e., severity) are given by:

INPUT

PROCESS

OUTPU T

I
I
i

DATA INTERACTIVE PRINT OUT REPORTS

Proba bilify =

Freauencu of relevant events Total number of possible events

I
RI! arcbirlCATION
POI-PICI

M AIN
PROCESSING OPTIONS

I

Amount at Stake = cost of investment loss + least cost to restore status quo
5. Overall Evaluation The third leve1 of risk modeling considers the impacts of ali risks combined, usually translated into project economics as a common basis. This approach is important since the economic return for a project is detennined from the scenarios of project costs, schedules, production, political considerations, and so on, over the whole life of the entity. Nevertheless, considerations of economics alone will not deal with such dimensions as safety and environmentalimpacts. Depending on the project, these may very well require independent presentation.
6. Project Risk Report

I

Ik,r

i

CONFIGURATION

CALCULATIONS

CONFIGURATIONS SPECIFICATION

RISK VALUES

CONSEQUENCES

ACROSS RISKS

The resulting report should be prepared as a baseline summarizing the findings, changes and/or recommendations, in a form that can be used for subsequent risk response planning, tracking and program updating. Each activity and associated risk events should be clearly described and numbered for referente, along with the impact scenarios envisaged. Some form of tabulation of the nsks considered and their comparative values showing status ranking will be helpful, as weli as examples of the major decision trees, and any schedule nsk networks developed. In its simplest form, the expected total value of risks on the project might be presented as a "bottom line" to the project costs as foliows:

m
TRANSFORMATION

Figure B.2. Simplified Risk Analysis Program Structure

After A.B. Carnrnaert c. 1986

1. Adapted from notes by Dr. A.B. Cammaert, whileManager, Arctic and Off-shore SeMces, Acres International Limited, Toronto. 2. D.T. Hulett, PMP Certification Workshop - Risk Management, PMI Orange County Chapter, 1991, p3. 3. J.R. Adarns, M.D. Martin, A Practical Approach to the Assessment of Project Uncertainty, P Seminar/Symposium Proceedings, Toronto, 1982, p IV-F.3. M

Appendix C

Appendix C
Other Risk Analysis Techniques
Some techniques which can be used to support risk analysis follow. Practical application may be limited to certain types or size of project. Perhaps more important is management's attitude towards risk analysis itself, especially as it tends to be governed more by their understanding of the mathematics involved in the techniques and, consequently, in their confidence in the results produced.
1. Brainstorming

This technique is used extensively in formative project planning, and can also be used to advantage to ident* and postulate risk scenarios for a particular project. It is a simple but effective attempt to help people think creatively in a group setting without feeling inhibited or being criticized by others. A committee is assembled, whose members have as relevant and broad a knowledge of the circumstances of the situation as possible. The rules are that each member must try to build on the ideas offered by preceding comrnents. No criticism or disapproving verbal or nonverbal behaviors are allowed. The leader or facilitator does not require any special expertise, but must enforce the mies and record the results. The intent is to encourage as many ideas as possible, which may in tum trigger the ideas of others. I assembling the ideas, any evaluation is strictly reserved for n later study. Thus, ideas are encouraged to flow as freely as possible, however imaginative, innovative or wild they may appear. While many suggestions may subsequently be rejected, the greater the number to start with, the more likely that a useful number will be retained, and so provide more comprehensive coverage. The technique is improved by the variety in the participants' backgrounds and is very helpful in project team building. It is very effective in finding creative solutions to potential problems.

2. Sensitivity Analysis
Sensitivity analysisl seeks to place a value on the effect of change of a singlevariable within a project by analyzing that effect s on the project plan. It i the simplest form of risk analysis. Uncertainty and risk are reflected by defining a likely range of variation

Other Risk Analysis Techniques
for each component of the original base case estimate. In practice such an analysis is only done for those variables which have a high impact on cost, time or economic return, and to which the project will be most sensitive. The effect of change of each of these variables on the final cost or time criteria is then assessed in turn across the assumed ranges. If several variables are changed, the most sensitive or critica1 variables can be compared graphically in a sensitivity diagram. Some of the advantages of sensitivity analysis include impressing management that there is a range of possible outcomes, decision making is more reahtic, though perhaps more complex, and the relative importante of each variable exarnined is readily apparent. Some weaknesses are that variables are treated individually, lirniting the extent to which combinations of variables can be assessed, and a sensitivity diagram gives no indication of anticipated probability of occurrence.

Appendix C
A scenario is established and each panelist is requested to reply to a questionnaire. The responses, together with opinions and justifications, are evaluated and statistical feedback is furnished to each panel member in the next iteration. The process is continued until group responses converge to a specific solution. Should the responses diverge, the facilitator needs to review the wording of the questionnaire, the feedback, or the experience of the panelists to determine if there is a problem which needs to be corrected. Social scientists are sometimes critical of the method on the grounds that the method has no predictive validity, and that the use of "experts" leads to manipulation of group suggestion rather than consensus.

5. Monte Carlo
The Monte Carlo ~ e t h o d ? simulation by means of random numbers, provides a powerful yet simple method of incorporating probabilistic data. The basic steps are:
I. Assess the range for the variables being considered and deter-

3. Probability Analysis
Probability analysis2overcomes the limitations of sensitivity analysis by specifying a probability distribution for each variable, and then considering situations where any or a11 of these variables can be changed at the same time. However, since every project is unique, defining the probability of occurrence of any specific variable may be quite difficult, particularly as political or commercial environments can change quite rapidly. Typically, a distribution profile is allocated to the range which has been defined for the variable, and, in the absence of statistical data, simple triangular, trapezoidal or rectangular distributions may be adopted. As with sensitivity analysis, the range of variation is subjective, but ranges for many time and cost elements of a project estimate should be skewed toward overrun, due to the natural optimism (or omission) of the estimator. The problem of assessing how risks can occur in combination is usually handled by a sampling approach (such as the Monte Carlo technique below) and running the analysis a number of times. The outcome is a range of possible results with their respective probabilities shown diagrammatically, such that reviewers can assess their own attitudes and response to the project and its nsks.
4. Delphi Method

mine the probability distribution most suited to each.
2. For each variable within its specific range, select a value

randomly chosen, taking account of the probability distribution for the occurrence of the variable. This may be achieved by generating the curnulative frequency curve for the variable and choosing a value from a random number table. 3. Run a deterministic analysis using the combination of values selected for each one of the variables. 4. Repeat steps 2 and 3 a number of times to obtain the probability distribution of the result. The number of iterations required depends on the number of variables and the degree of confidente required, but typically lies between 100 and 1000.

The basic concept of the Delphi ~ e t h o is ~ derive a cond to sensus using a panel of experts to amve at a convergent solution to a specific problem. This is particularly useful, for example, in amving at probability assessments relating to future events where the risk impacts are large and critical. The first and vital step is to select a panel of individuais, as participants, who have expenence in the area at issue. For best results the panel members should not know each othersidentity,and the process should be conducted with each at separatelocations. T i is to prevent singlemember influente hs and simplistic concurrence.

6. Decision Tree Analysis A feature of project work is that a number of options are typically available in the course of reaching the final results. Indeed, even before considering the project in any detail or developing a network analysis for example, the decision maker is faced with an array of procurement possibilities and a sequence of e ~ decision choices. The Decision ~ r e provides a graphical means of bringng the information together. Figure C.l shows the principle applied to the choice between two projècts. An advantage in its application to risk rnanagement is that it forces consideration of the probability of each outcome. Thus, the likelihood of failure is quantified and some value is placed on each decision. This form of nsk analysis is usually applied to cost and n time considerations, both i choosing between different early investment decisions, and later in considering major changes with uncertain outcomes during project irnplementation. In the latter case, it may be linked to a sensitivity analysis as a means of determining the value of a certain decision.

Other Risk Analysis Techniques

SUCCESS -30 PROJECT A FAILURE START
(-28)
(024) 020

SUCCESS PROJECT B FAILURE

035 d

(-12) 015

Note the change in recommendation if the initial probability ( ) is changed

Figure C.1. Decision Tree for Two Projects Showing Probabilities Assigned
After J.R. Adams and M.D. Martin, A Practical Approach to the Assessment of Project Uncertainty, PMI Seminar/Symposium, Toronto. 1982, plV-F.7

The method is well suited to project risk analysis and has been applied extensively, with additional efíorts made to resolve the prob lem of interrelated risks.

7. Utility Theory
None of the techniques discussed so far take into account the attitude towards risk of the decision maker. It may be reasonable to suppose, for example, that a potential loss of 90 percent would not be viewed with the same equanirnity as, say, a loss of 10 percent. Somewhere in between the attitude will change. However, at what point may weii depend on the attitude of the decision maker. That is to say, the decision maker may be risk seeking, risk neutral, or risk averse. Uality Theory6 endeavors to formalize managementfs attitude towards risk, an approach which is appropnate to Decision Tree Analysis for the calculation of expected values, and also for the assessment of results from sensitivity and probability analyses. However, in practical project work Utility Theory tends to be viewed as rather theoretical.

8. Decision Theory
Decision Theory7 U a technique for assisting in reaching decisions under uncertainty and risk. A decisions are based to U some extent on uncertain forecasts. Given the criteria selected by the decision maker, Decision Theory points to the best possible course whether or not the forecasts are accurate.

1. J.G. Perry and R.W. Hayes, Risk and its Management in Construction Projects, Proceedings of the Institute o Civil Engineers, Part 1,1985, June 1985. f 2. Ibid. 3. M.D. Martin and M.B. McCormick, Improving Project Planning Productivity, PMI Seminar/Symposium,Houston, Texas, 1983, p I11 E5. 4. J.G. Perry and R.W. Hayes, Risk and its Management in ConstructionProjects, Proceedings o the Institute of Civil Engineers, Part 1,1985, June 1985. f 5. Ibid. 6. Ibid. 7. After J.R. Adams and M.D. Martin, A Practical Approach to the Assessment of Project

Appendix D

Appendix D
Risk Applied to Schedule and Cost Analysis
Optimistic-Pessimistic Estimating It is often possible to describe the likelihood of particular events occurring in such vague t e m as quite likely, often, rarely, etc. However, for project purposes it is typically necessary to quantify the probability of an event occurring in order to plan an appropriate response. The use of statistical analysis tools, whether they result in subjectiveor objective probabiiities, enables quantification with some degree of confidente. For example, in order to develop a realistic project schedule and/or cost, it is necessary to know the durations and/or cost of the various activities involved. When some of those activities are very uncertain, perhaps because they are unfamiliar, a different strategy is required. One estimating strategy which is useful in these circumstances requires making three estimates. The following approach can be appiied to either cost estimating or to time estimating, but for simplicity, the method will be described in the context of scheduiing. The first estimate is an optimistic-corresponding to the shortest possible time if everything falls neatly into place as required. The second estimate is the most likely time-one in which there is a typical balance between things going well and things going poorly.. The third estimate is a pessimistic o n e t h e time it wili take if difficuities are encountered and a lot of things go wrong. These three estimatescan be more technicaliy described as follows:
Optimistic time estimate (o): A time estimate in which a task can be completed if everything goes exceptionaüy well. An estimate in which the probability of accomplishrnent is not more than onein-one hundred (1%), the original PERT assumption. Most likely time estimate (m): The estimate of the time required for an activity which would be expected to occur most often if the activity could be repeated severa1 times under identical circumstances (without any "learnhg curve" effects). Pessimistic time estimate (p): An estimate of the longest time a task rnight require under the most adverse conditions, baning "acts of God." An estimate i which the probabiiity of not accompn lishing the task is less than 1%.

Risk Applied to Schedule and Cost Analysis
Note that the most-likely-time is a matter of judgement. It is not the same as the "expected time." Expected time (te) is a term given to a calculated weighted average calculated as foliows: Expected Time (te): An estimated value calculated from the formula k = (0+4m+p)/6, where o = optimistic, p = pessimistic and m = most-likely. This relationship is a simplification of the "bellshaped" curve of probability theory, giving recognition to the fact that such time estimates generally are not symrnetrical. In amving at the three judgrnental estimates, it is desirable to consult with members of the project team who have relevant experience of each activity in question. It will not only lead to establishingbetter values, but, as a side benefit, involves members of the team and thereby buiids cornmitment to the project plan. Consider the project plan iilustrated below in activity-onnode notation with one time estimate for each activity. Conventional critica1 path calculations lead to an indicated completion time of 30 days.

Appendix D
Again, the indicated completion of the project is 30 days but that is now called the "expected completion" of the project. There is a 50% chance of exceeding that time as well as a 50%chance of completing before that time. Using a concept from probability theory, the Central Limit Theorem, we can combine the uncertainty about the individual activities into an expression of the uncertainty about the project completion time. This requires the calculation of the standard deviations of each activity using the formula, standard deviation = (p - 0)/6, per the original PERT concept. Thus, the standard deviations for activities A, C, and F are 2.00 days [(I5 - 31/61 and for activities B, E, and G, 0.33 days [(I1 - 91/61. For activity D it is, of course, zero. To get the standard deviation of the completion of the project, the uncertainty of the activities on the critica1 path must be combined. Theory requires that the standard deviations be "squared" to obtain the "variances" in order to add them so the variances of the activities B, E, and G are 0.111. Adding them gives a variance of .333 of which the "square root," i-e., the standard deviation, is 0.577 days for project completion. This standard deviation can be used with the standard normal distribution to make statements about the probability of completing the project by a specific time. For exarnple, the 95%confidence lirnits on the normal distribution is based on plus or rninus 1.96 standard deviations. Multiplying 1.96 times .577 days gives 1.13 days from which we can say that there is a 95% probability of completing the project in between 28.87 days and 31.13days. If the target completion date is 31 days, we can state that the probability of being completed by that date is 96%.Thus, the risk of being late or early can be assessed. This illustration was designed to emphasize a fallacy of the original PERT concept. It assumed that the uncertainty of project completion was dependent only upon the critica1 path. In this illustration, it can be seen that the second most critica1path has a duration of 27 days. However, it is composed of activities which are far more uncertain. They each have standard deviations of 2 days resulting in a standard deviation for the completion of that path being 3.46 days. Thus, the 95% confidence lirnits for the completion of the project, based on this path, are 20.22 days to 33.78 days and the probability of project completion in 31 days is now only 88%. This path has greater potential of causing a delay in project completion than the critical path and therefore poses greater risk. Asolution to this anomoly exists however, in project management software which incorporates a Monte Carlo Simulation approach to analyzing uncertainty. This approach is identical to the above up to the point of calculating the project completion time. Instead of using the expected times (te), a procedure is used which takes samples from the range of durations for each activity using random numbers. Thus, in any given calculation, one activity may have a performance time near its optimistic, another near its pessimistic, and another near its most likely, and so forth. The

Suppose estimates were obtained for these same activities using the three time estimate approach and the assumptions in the original PERT system. The network and estirnates rnight be as shown below and result in the tes as shown. Note that these estimates are probably not realistic but are chosen to simplrfy the iílustration and emphasize the a dificiency in conventional PERT.

Risk Applied to Schedule and Cost Analysis
frequency of selection of times within this range is consistent with the probability distribution of durations for each activity. The resulting completion time from that calculation is then an observation in the development of the distibution of completion times for the project completion and is a function of the uncertainty of every activity in the project. This resulting distribution is then a reasonably accurate reflection of the uncertainty associated with the completion of the project and can be used effectively in risk analysis of project completion times. There have been several variations developed from the original PERT. One of those, by ~ o d e r ' ,argues that the optimistic and pessimistic time estimates are more likely to be a reflection of one-in-twenty occurences than one-in-one hundred. If that is true, then the divisor for the standard deviation is 3.2. Othenvise the calculations and concepts are the same.

Appendix D
developing resuiting scenarios. Expected Value (EV) can be used to adjust the value of the consequences of any given outcome for the probabiiity of its occurence. Assume that our project has an estirnated cost of $90,000 and has to be completed in 31 days. In addition, there is a $50,000 penalty if completion takes longer. How significant is this risk? In our schedule example described earlier, based on the most risky path there was an 88% chance of completing the project in 31 days or less; therefore, there is a 12% chance that the project would not be completed in 31 days. The expected cost is calculated as the sum of the products of the value of an outcome times the probability that that outcome will occur as follows.
Outcome Complete in 31 days or less Complete in more than 31 days Value of the Outcome $90,000 140,000 Probability Product .88 .12 Expected Value
X

x

-

-

$79,200 16,800 $96,000

Impact Analysis: Probability of Event and Severity of Consequences The relationship between probability and severity of consequences is shown schematically in Figure D.1. Event probabilities can often be estirnated using statistical inferences based on history. Severity of consequences may be similarly derived or by estimating the impact of specific events by

PROJECT R I S K S

RECURRING CONDITIONS

NON-RECURRING EVENTS

OBJECTIVE ANALYSIS

SUBJECTIVE ANALYSIS

From this calcuiation it will be seen that the EV of the cost is substantially higher' than the estimated cost and the calculation provides a clearer basis for management decision as, for example, a decision as to whether or not to invest in an R&D project. The calcuiation also provides a basis for other comparative calcuiations. For instance, whether or not a scheduie slippage beyond the 31 day point is a high risk ora low one depends on the consequences that rnay result. In rea1:world t e m , it may also depend on who is responsible for the slippage. Consider the "normal" experience that design changes cause schedule slippages. Suppose that our experience of the type of project calculated earlier shows that there is a 50/50 chance of the project's client requesting a design change, and further that according to our records client-ordered changes have led to a delay on 70% of the occasions. Then, if a schedule slippage occurs on this project, what is the likeiihood that it wiil be caused by a client-ordered design change? This calculation is shown below:

p%%%%%mln%m%%t%"n-.,

$ (PRIORITY) 1
*
2
%=ly=nnnp=swSnnp-*

HIGH RISK

i

-5 x . = .35 7
delay
Pclient

PROBABILITY OF OCCURRENCE

= .35/.41 = -85

LOW

I

1
SE VERTTY OF CONSEOUENCES

I
change de lay
Pnot ciient = .06/.41 = 15

.5x .12= -06 -7'7-

7%r

RISK CRITERION VALUES (RISK EVENT STATUS)
Figure D.1.

no delay .8

0

Risk Applied to Schedule and Cost Analysis
The results show that there is an 85% chance of the responsibiiity being with the client, and therefore a 15% chance that the slippage w i l l m be due to the client. The calculation of the revised EV of cost is shown as $90,900.

Appendix D
expected completion time of 30 days (derived from the original PERT calculations) or savings from completing in less than 30 days, outcomes which each have a probabiiity of 50%. The EV of the profit from a FFP of $115,000, with penaity, is $24,000. A risk taker could be expected to take the second option as the EV of profit is greater. In other words, there is a 98% probability of a $25,000 profit even when offset by an 2% chance of a $25,000 loss. This could be considered a better deal than a $10,000 profit guaranteed. Note, however, that expected value theory assumes that the risk taker is playing this game a large number of times and that in any single play the risk taker can afford to take the loss, should it occur.

.12 x .85 x $90,000 = $9,180 >31 days .12

not client .15

O'

. 1 2 x . 1 5 ~ $ 1 4 0 , 0 0 0 = $3,520 Expected Value =

Taking into account that there is only a 12% chance of overrunning the 31 day limit and only a 15% chance of the penalty being imposed, and that these two events are assumed to be independent, then the probability of incurring the $50,000 late completion penaity is about 2% (12% x 15%). Conversely, the probability of there being no penalty, even if the project is late, is 98%.

REVENUE W/Openalty clause $100,000

less

EXPECTED COST

=

EV PROFIT

$1 15,000 W/ penalty clause

cPog8
0
1.O x $90,000 = $90,000 Expected profit = $1 0,000

versus

x $ 90,000 = $88,200

.O2 x $140,000 = $2,800 Expected cost = $91,000 Expected profit = $24,000

Guidelines for Use of Expected Value Techniques 1. The risk assessment process should not be started with a preconceived point of view. 2. Keep an open mind until an objective decision has been reached. 3. Ensure objectivity in defining the risk, reward and remedy and in the data coilection process. 4. Avoid biased subjectiveprobabiiity judgements and accurately estimate the value of thereward and the cost of the potential remedial action. 5. If the probabilities are defined by a range of values, use n the conservative end of the range i the analysis. 6. If the computed expected vaiue of profit is clearly negative, DON'T TAKE THE RISK, if that is an option. 7. If the computed expected value of profit is clearly positive, TAKE THE RISK, provided the loss can be afforded, should it ocm. 8. If the sign (positive) of the computed expected vaiue of profits can change with siight adjustments in the probability n judgements or i the estirnated reward or remedy, DON'T TAKE THE RISK. 9. If there are .any doubts at ail about the situation (risk, reward, remedy or probabfities of success and failure), DON'T TAKE THE RISK. 10. Once the risk has been taken, be ready to implement a contingency plan if and when it is realized that the desired outcome cannot be achieved.

Consider the possibility of having a choice between taking on the project at a Firm Fixed Price (FFP) $115,000 with the penalty of clause, or at $100,000 without the penalty clause. Which option should be taken? An aid to answering this question can be obtained by using the above probabilities to compare the two alternatives, FFP without penalty clause and FFP with penalty clause. The probabfity calculation shows that the EV of the profit from a FFP of $100,000 with no penalty is $10,000. This does not consider any increase in costs due to the project exceeding the

1. Moder, Joseph J., Ceci1 R. Philiips, and Edward W. Davis. 1983. Project Management With CPM, PERT and Precedente Diagramrning. Third Edition. p. 283. Van Nostrand Reinhold Company: New York

D-7

Appendix

E

Appendix E
A Glossary of Project and Program Risk Management Terminology
Note: The project and program rlsk management context 1 Impilclt s throughout the following definitlons.

Activities. A senes of tasks performed over a penod of time. Amount at Stake. The extent of adverse consequences which could occur to the

project.
Business Risk. The inherent chances for both profit or loss associated with a partic-

ular endeavor.
Contingency Allowance. Specific provision to cover variations which may occur in the

expected values of elements of cost or schedule,but not scope or quality (see Contingency Reserve).
Contingency Plannlng. The development of management plans to be invoked in the

event of specified rzkk events. Exarnples include the provision and prudent n management of a contingency allaoance i the budget, the preparation of alternative schedule activity sequences, work-arounds and emergency responses to reduce the impacts of particular risk events, and the evaluation of Liabilities in the event of complete project shut down.
Contingency Reserve. Aprovision held by the project sponsor for possible changes in

project scope or quality. Scope and quality changes constitute changes in the project manager's mandate and will affect the project's cost and schedule (see Contingency Allowance).
Control. Planning, monitoring accomplishment, and exercising any necessary cor-

rective action to yield the required outcome.
Deflection. The act of transferring all or part of a risk to another party, usually by some

form of contract.
GolNo-Go Decision. A major decision point in the project life cycle, typically marking

the transition from planning to accomplishrnent.
Historical Database. Records accumdating past project expenence stored as data for

use in estimating, forecasting and predicting future events.
Insurable Risk. A particular type of nsk which can be covered by an insurance policy.

A Glossary of Project and Program Risk Management Terminology
Impact Anaiysis. The mathematical examination of the nature of individual risks on Risk Factor. Any one of risk event, risk probabilify or amount at stake.

Appendix E

the project, as well as potential arrangements of interdependent risks. It includes the quantification of their respective impact severity, probability, and sensitivity to changes in related project vanables, including the project life cycle. To be complete, the analysis should also include an exarnination of the externa1 "status quo" prior to project implementation as well as the project's interna1 intrinsic worth as a reference baseline. A determination should also be made as to whether all risks identified are within the scope of the project's response planning process.
Known, Known-Unknown, Unknown-Unknown. A method of classifymg risks according

Ri~k identification. The process of systematically identlfying all possible risk events

which may impact on a project. They rnay be conveniently classified according to their cause or source and ranked roughly according to ability to manage effective responses. Not a11 risk events will impact all projects, but the cumulative effect of several risk events occuning in conjunction may well be more severe than exarnination of individual risk events might suggest.
Risk Management. The art and science of ident%ng, analyzing and responding to

to the amount of information available.
Mitigatlon. The act of revising the project's scope, budget, schedule or quality,

risk factors throughout the life of a project and in the best interests of its objectives.
Risk Probability. The degree to which the risk event is likely to occur. Risk Response System. The ongoing process put in place during the life of the project

preferably without material impact on the project's objectives, in order to reduce uncertainfy on the project.
Opportunity.The cumulative effect of the chances of uncertain occurrenceswhich will

affect project objectives positively. Opportunity is the opposite of risk.
Post-Project Review. An appraisal of all aspects of a project upon completion, with a

to monitor, review and ugdate projed risk and make the necessary adjustments. Exarnination of the vanous risks will show that some risks are greater in some stages of the project life cycle than in others.
SUCC~SS (Project Success). The achievement of stakeholder satisfaction. Surprise. The surfacing of an unantiapated uncertainfy,either apportunity or risk event. Technique. Skilled means to an end. Total Certainty. A 1 information is known. 1 Total Uncertainty. No information is available and nothing is known. By definition,

view to examining and documenting variations and events, to augment the organization's historical database.
Probability.The likelihood of occurrence.The ratio of the number of chances by which

an event may happen (or not happen) to the sum of the chances of both happening and not happening.
Process. The set of activities required to achieve an output. Pure Risk. See insurable risk Publlc Relations.An activity designed to improve the environment in which a project

total uncertainty cannot be envisaged.
Uncertainty.The possibility that events may occur which will impact the project either

organization operates in order to improve project performance and reception.
Response Planning. The process of formulating suitable risk managernent strategies for

favorably or unfavorably.Uncertainty gives rise to both opportunity and risk.
Work-around. An alternative solution to a potential problem.

the project, including the allocation of responsibility to the project's various functional areas. It may involve mitigation, deflection and contingency planning. Some flexíbility should alço be provided, however tentative, for the completely unforeseen occurrence.
Risk (Project Risk). The cumulative effect of the chances of an uncertain occurrences which will adversely affectproject objectives. It is the degree of exposure to negative events and their probable consequences. Project risk is character-

ized by three risk factors: risk event, risk probabilify and the amount at stake. Risk is the opposite of opportunity.
Risk Data Applications. The development of a database of risk fadors, actual responses

and consequences, both for the current project and as a matter of historic record.
Risk Event. The precise description of what might happen to the detriment of the

project.
Risk Event Status. Ameasure of importance of a risk event. Also referred to as criterion

value or simply its ranking.

YOUR COMMENTS
on

Project and Program Risk Management
HELP US HELP YOU!
So that we rnay better provide you with the practical information you need, please take a rnoment to record your cornments on this page and return it to:

The Executive Director
Project Management Institute
130 South State Road

Upper Darby, PA 19082

U.S.A.
I/we have the following suggestions regarding this handbook:

Address City Country State

ZIP Code
Date

Code of Ethics for Thc Project Management Profession PREAMBLE: Project Management Professionals. in lhe pursuit of the profession. affect the quality of life for all people in o c r society. Thcrefore. it is vital that Project Management Professionals conduct their work in an ethical manner t o earn and maintain the confidence of team mernbers, colleagues. ernployees, employen, ciients and the public. ARTICLE I: Project Management Professionals shall rnaintain high standards of personal and professional conduct. and: a.
b.
c.

II
R. M a x Wideman,P.Eng. FEIC, FICE, Fellow PMI, is a professional engineer specializing in project management consulting. Since graduating at London University,his e v ' e n c e has included hydroelectric, river, marine, transportation, industrial, institutional, wmmercial and residential projects. He has also been instrumental in social and environmental impact studies, major contract and expropriation claims, wnstruction productivity, and project management audit. In working for a diversity of sectors, he has gained a broad perspective and insight into the project management process. Mr. Wideman has lectured extensively, presenting papers or seminars on a variety of project management topics in Canada, China, Egypt, Iceland, India, lamaica, Pakistan, the Philippines, Saudi Arabia, the United Kingdom, and the USA. In 1974, Mr. Wideman joined the Project Management Institute (PMI) and later launched the PMI West Coast BC chapter. In 1982 he was elected to the International Board as Vice President Member Services and s d as director for three years. During this time, he was assigned responsibilityfor expanding and codifying PMIS existing standards of knuwledge by conducting a major voluntary study by PMI members. The resulting report became known as the Project Management Body of Knowledge, or "PMBOK," which m s approved by the PMI Board in March 1987. Mr. Wideman received PMI's Distinguished Contribution to Project Management Award in 1985, and the following year was honored as PMI Personof-the-Year. He zoas elected PMI president for 1987, became chaiman in 1988, and was made a Felluw of the Institute in 1989. Mr. Wideman has authored a number of articles and papers for the Institute's publications and is author of C o ç t o l of AE W Semices, Vancouver, 1983.

Accept responsibility for their actionc Undenake projects and accept responsibility only if qualified by training o r experiente. or after full disclosure to their employers o r clients of peninent qualifications. hlaintain their professional skills at the state of the art and recognize the imponance of continued personal developrnenr and education. Advancc the intrgrit? and prestige of the profession bg practicing in a dignified manner. Suppon this code and encounge colleapues and co-workers to act in accordance with this code. Suppon ihe professional society by actirelg participating and encounping eolkagues and co-workers to participate. Obek the laws of the country in which work is being performed

II I/

I/

II
II

II

d

e.

II

I.
.

II

II

A RTICLE I I : Project Management Professionals shall. in their work: a. Pro\.ide the necessary project leadership to promote maximum productivity while striving to rninimize C O S ~ S . .4ppI! state of t he art project rnanagement toolsand techniques t o ensure quality. cost and time objectives. as set fonh in the project plan. are rnet. Treat fairly all project team members. colleagues and co-workers. regardlcss of race. religion. scx. age or national origin. Protect project tearn members frorn physical and mental harm. Provide suitablc working conditions and opportunities for project team members. Seck. accept and offer honest criticism of work, and properly credit the contribution of othcrs Assist project tearn members, colleagues and co-workers in their professional development. Act as faithful agents o r trustees for thcir employers and clients in professional o r business matters. Keep information on the business affairs o r technical processes of a n employer o r client in confidence while employed. and iater. until such information is propcrly relcased. Inform their employers. clients. professional societies o r public agencies of which they are members o r to which they may make any presentations. of any circumstance that could lead t o a conflict of interest. Neither give nor accept. directly o r indirectly. any gift, payment o r service of more than nominal value to or from those having business relationships with their employers o r clients.
E honest and realistic in reporting project quality. cost and time. k

I
II

II

/I
II
II

b.
c. d. e f. g. a. b. c. d. e. a. b.

II

ARTICLE 111: Projcct Management Professionals shall, in their relations with employers and clients:

ARTICLE IV: Project Management Professionals shall. in fulfilling their responsibilities t o the community: Protect the safety. health and welfare of the public and spcak out against abuses i? these arcas affecting lhe public interest. Seek t o extend public knowledge and appreciation of the project management profession and its achievements.

r


								
To top